Skip to main content

Full text of "Journal of the Society of Motion Picture Engineers"

See other formats

From the collection of the 



o Prelinger 

i a 


t w 

San Francisco, California 


Vol 45 JULY, 1945 No. 1 


Comparison of Variable-Area Sound Recording Films 

D. O'DEA 1 

Method for Measurement of Brightness of Carbon Arcs 


Frequency Allocations for Theater Television : Excerpts 
from Report by Federal Communications Commission 
on Proposed Allocations from 25,000 Kilocycles to 
30,000,000 Kilocycles 16 

The Application of the Polarograph to the Analysis of 
Photographic Fixing Baths 


Progress Report of the Work of the ASA War Committee 
on Photography and Cinematography, Z52 

J. W. McNAiR 33 

An Automatic High-Pressure Mercury Arc Lamp Con- 
trol Circuit L. F. BIRD 38 

Anecdotal History of Sound Recording Technique 


A New Medium for the Production of Vandykes 


Current Literature 65 

Society Announcements * 67 

Program of 57th Semi-Annual Technical Conference 68 

Copyrighted, 1945, by the Society of Motion Picture Engineers, Inc. Permission to republish 
material from the JOURNAL must be obtained in writing from the General Office of the Society. 
The Society is not responsible for statements of authors or contributors. 

Indexes to the semi-annual volumes of the JOURNAL are published in the June and December 
issues. The contents are also indexed in the Industrial Arts Index available in public libraries. 





Board of Editors 





Officers of the Society 
**President: DONALD E. HYNDMAN, 

350 Madison Ave., New York 17. 
** Past-President: HERBERT GRIFFIN, 

133 E. Santa Anita Ave., Burbank, Calif. 
** Executive Vice-President: LOREN L. RYDER, 

5451 Marathon St., Hollywood 38. 
^Engineering Vice-President: JOHN A. MAURER, 

37-01 31st St., Long Island City 1, N. Y. 
** Editorial Vice- President: ARTHUR C. DOWNES, 

Box 6087, Cleveland 1, Ohio. 
^Financial Vice-President: ARTHUR S. DICKINSON, 

28 West 44th St., New York 18. 
** Convention Vice-President: WILLIAM C. KUNZMANN, 

Box 6087, Cleveland 1, Ohio. 
* Secretary: E. ALLAN WILLIFORD, 
230 Park Ave., New York 17. 
^Treasurer: M. R. BOYER, 
350 Fifth Ave., New York 1. 


*FRANK E. CARLSON, Nela Park, Cleveland 12, Ohio. 
**JOHN I. CRABTREE, Kodak Park, Rochester 4, N. Y. 
**CHARLES R. DAILY, 5451 Marathon St., Hollywood 38. 

*EDWARD M. HONAN, 6601 Romaine St., Hollywood 38. 
"{CLYDE R. KEITH, 233 Broadway, New York 7. 

*G. T. LORANCE, 92 Gold St., New York 7. 
**PETER MOLE, 941 N. Sycamore Ave., Hollywood. 
*fHoLLis W. MOYSE, 6656 Santa Monica Blvd., Hollywood. 
** WILLIAM A. MUELLER, 4000 W. Olive Ave., Burbank, Calif. 

*EARL I. SPONABLE, 460 West 54th St., New York 19. 
**REEVE O. STROCK, 111 Eighth Ave., New York 11. 

*WALLACE V. WOLFE, 1016 N. Sycamore St., Hollywood. 

Term expires December 31, 1945. tChairman, Pacific Coast Section. 
**Term expires December 31, 1946. ^Chairman, Atlantic Coast Section. 

Subscription to nonmembers, $8.00 per annum; to members, $5.00 per annum, included in 
their annual membership dues; single copies, $1.00. A discount on subscription or single copies 
of 15 per cent is allowed to accredited agencies. Order from the Society at address above. 
Published monthly at Easton, Pa., by the Society o'f Motion Picture Engineers, Inc. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 

General and Editorial Office, Hotel Pennsylvania, New York 1, N. Y. 

Entered as second-class matter January 15, 1930, at the Post Office at Easton, 

Pa., under the Act of March 3, 1879. 


Vol 45 JULY, 1945 No. 1 




Summary. This paper describes the test results obtained by comparing the new 
Eastman 1372 film with those in current use. Our tests indicate that this film has 
characteristics superior to the Eastman films now in use for variable-area sound re- 
cording, particularly with respect to density speed, processing tolerances, and require- 
ments for direct positive. 

A new film, Eastman Fine-Grain Sound Recording Film, type 1372, 
has been placed on the market for use in variable-area sound record- 
ing. Our tests show that the characteristics of this film are superior 
to the Eastman films now in use for this purpose. 

The commercial value of a film can be indicated by exposure tests, 
sensitometric data, high-frequency attenuation measurements, and 
cross-modulation tests. 1 Other practical considerations which are of 
importance, such as aging, durability, freedom from fog, and uni- 
formity, can only be found after extensive use in the field, and are 
not within the scope of this paper. 

The cross-modulation tests, in brief, consist in making recordings of 
1000 cycles and 9000 cycles modulated by 400 cycles exposed with a 
wide range of lamp currents. The high-frequency attenuation 
measurements are made by recording 9000 cycles at the same time. 
This negative is developed and then printed at different print densi- 
ties to obtain a family of curves. The 1000-cycle recording serves as 
a reference level, and the output of the 9000 cycles indicates the loss at 
high frequencies. The 9000 cycles modulated by 400 cycles are re- 
produced through a 400-cycle band pass filter, and the combination 
of negative and print density which cancels the image spread .will 

* Presented Oct. 17, 1944, at the Technical Conference in New York. 
** RCA Victor Division, Radio Corporation of America, Hollywood. 


Vol 45, No. 1 


-50 ' 5*2 ' 5*4 ' 5*6 ' 5 S ' 5~0 52 ' 5* 65~ SB ^HT 


FIG. 1. Exposure tests, lamp current versus density. Developed in 
variable-area negative developer. 













^ - 





















.1 6 


2.0 2.4 2.8 


FIG. 2. Processing characteristics, 5372. Nega- 
tive exposed with ultraviolet light and developed in 
variable-area negative developer. 


give minimum output, as the 400-cycle component is only present 
when the printer, exposure, and developing conditions are not bal- 
anced and the average transmission changes at a rate of 400 cycles. 
It has been found in past experience that cancellation of 30 db or 
more is satisfactory. 

To assure accurate results, the equipment employed for this series 
of tests was first carefully aligned. The recorder was not refocused 
between unfiltered light and ultraviolet light recordings. However, 
when it was learned that the 1372 is 0.0003 in. less thick than other 
variable-area sound recording stocks, the recorder was refocused 
for the new film. A negative and prints were made which compare 
closely with those made before refocusing. 

Comparisons were made between the new film, 1372, and the 1357 
and 1 302 films. The 1357 has been used successfully as a variable- 
area sound negative material since 1936, and the 1302 has been used 
as a negative and print stock. The 137 2 -type emulsion which was 
used for our tests was on acetate base cut to 35-mm width for test 
samples, and therefore bore the code number 5372. When the regular 
1372 on nitrate base was available, more tests were made to check the 
new stock and the only difference found was a drop in the negative 
balance density from 1.3 to 1.2. The 1372 film has a blue-gray base 
with a density of approximately 0.24 (read on a visual densitometer), 
which is a lower density than the 5372. All negative density read- 
ings mentioned in this paper include the base density. When using 
pieces of "fixed out" film in the optical path of the reproducer, the 
loss in output due to the base density of the film is 2 db fcr the 1372 
as compared with 0.5 db for the 1302. It is probable that the discrep- 
ancy between the density measurements and the loss in level is caused 
by the color characteristic of the base and the spectral sensitivity of 
the measuring equipment. One of the studios reports no noticeable 
difference in level when splicing the 1372 with a clear base stock such 
as 1302 and using them as print stocks for dailies. In general, it is 
planned to use only 1372 for dailies when the change-over to the new 
stock is complete in a studio, in which case there should not be any 

The high-density speed of the 1372 film makes it possible to expose 
the negative with ultraviolet light at reasonable values of lamp cur- 
rent, as shown in Fig. 1 . A Corning 597 filter 3-mm thick was placed 
in the optical path of the recorder for the 1372 and 1357 negatives. 
The 1302 film was exposed with unfiltered light. 


Vol 45, No. 1 

Prints which were to be developed in the release print developer 
were exposed with unfiltered light, and prints which were to be de- 
veloped in the variable-area negative developer were exposed with 



1000 CY 




1.6 2.0 2.4 2.8 


FIG. 3. Processing characteristics, 1357. Negative exposed with 
ultraviolet light and developed in variable-area negative developer. 


p o 




^ , 

O -a 




^ r 

Z -12 

a0 0< 





3 ifi 




OL l6 


D -?0 


O 20 

















FIG. 4. Processing characteristics, 1302. Negative 
exposed with unfiltered light and developed in variable- 
area negative developer. 

ultraviolet light. All prints were made on 1302 stock and printed 
on a modified Bell and Howell printer. 

The variable-area negative developer mentioned above has the 
characteristics of higher contrast, higher density speed, and higher 




shoulder break than the release print developer. The release print, 
or positive, developer fills the contrast requirement for picture 
prints. When the composite release print is developed in this solu- 
tion, the lower contrast of this developer is not ideal for variable- 
area sound tracks, and the lack of sharpness introduced by this condi- 
tion must be compensated for by the use of a high negative density. 

The gamma of the 1372 is appreciably higher than the 1357 and 
slightly higher than the 1302. 
When exposed with unfiltered 
light and developed in the 
variable-area negative devel- 
oper, the gamma is approxi- 
mately 3.6 as compared with 
2.9 for the 1357 and 3.5 for the 
1302. When exposed with 
ultraviolet light and developed 
under the same conditions, the 
1372 gamma is 2.95 as com- 
pared with the 1357 gamma of 
1.95 and 1302 gamma of 2.2. 
Detailed sensitometric data are 

covered in a paper entitled 
"Two New Fine-Grain Sound 
Recording Films" by R. M. 
Corbin, N. L. Simmons, and 
D. E. Hyndman.* 

An outstanding characteristic 
of this 1372 film, which is evi- 
dent in Fig. 2, is its ability to 
be used as a direct positive 
material. This negative de- 
veloped normally has very good high-frequency response, and the 
negative balance density, or density at which the image spread 
is minimum, occurs at the reasonably high density of 1.2. Pre- 
vious films have had negative balance densities no higher than 0.8, 
which resulted in low output and high noise level. This improvement 
is noticeable by comparison with the 1357 and 1302, as illustrated 
in Figs. 3 and 4. It is hoped that the advantages of the direct posi- 




I 8 





2.4 26 2fl 


FIG. 5. Processing tolerances, 5372. 
Negative exposed with ultraviolet light 
and developed in variable-area negative 
developer. Prints developed in positive 

* To be published in the JOURNAL. 


Vol 45, No. 1 

live method of recording for special purposes can be fully realized 
now that a suitable film is available. 

Families of prints made from negatives on these 3 films may be 
compared to illustrate the differences which will be found in practice. 
Prints were made in the release print developer, and summary curves 
from these families are shown in Figs. 5, 6, and 7. These summary 
curves are made from the cross-modulation tests by plotting the nega- 
tive densities which will give at least 30 db cancellation for different 

Z 16 


20 22 24 26 



FIG. 6. Processing tolerances, 1357. Negative ex- 
posed with ultraviolet light and developed in variable- 
area negative developer. Prints developed in positive 

values of print density. Although the cancellation would be equally 
good for almost any combination of negative and print density on 
these curves, it is necessary to select a print density of 1.3 or higher 
and use the negative density which corresponds. This value of 1.3 or 
higher is limited in the low density direction by the necessity of se- 
curing sufficient output and freedom from noise. It is limited in the 
high density direction by the high-frequency attenuation and nega- 
tive exposure available. 

In the case of the 1372, the negative density which gives the best 









2.0 ~22 24" 2.6 2 a 


FIG. 7. Processing tolerances, 1302. Negative 
exposed with unfiltered light and developed in 
variable-area negative developer. Prints developed 
in positive developer. 





































14 16 1.8 2.0 2.2 2.4 26 2.8 3.0 


FIG. 8. Processing tolerances, 557^. Negative exposed 
with ultraviolet light. Negative and prints developed in 
variable-area negative developer. 



Vol 45, Xo. 1 

cancellation at a normal print density is approximately 2.7. It will 
be noted from the summary curves that the tolerance in negative 
density which can be allowed for any set values of tolerance in print 
density is much greater for the 1372 than for the other 2 films. 

The high-frequency response of the 1372 is slightly better than that 
of the 1357 or 1302. Under normal conditions, that is, using the 
1357 as a negative developed in the negative developer and printed 
on 1302 stock developed in the positive developer, the attenuation at 
9000 cycles has been approximately 5 db. When the 1372 is used 

under the same conditions, the 
9000-cycle attenuation is only 

In order to get the best 
quality on prints from the origi- 
nal negative made for rerecord- 
ing or "dubbing" purposes, 
they are developed in the 
variable-area negative de- 
veloper. Families of prints 
were made under these condi- 
tions. From a comparison of 
these 3 curves, 8, 9, and 10, it 
will be noted that there is also 
more tolerance in the 1372. 
Under dubbing print conditions 
an improvement of 2 db in the 
high-frequency response is noted 




























*~ i/i 



z 14 












1.6 1.8 2.0 2.2 



FIG. 9. Processing tolerances, 1357. 
Negative exposed with ultraviolet light. 
Negative and prints developed in vari- 
ble-area negative developer. 

similar to that found under re- 
lease print conditions. It is ob- 
vious that the use of the 1372 
as the print stock would further 
widen the cross-modulation tolerances and improve the high-fre- 
quency response. 

Owing to the high resolving power of this new film it was considered 
possible that the negative could be developed satisfactorily in the 
print-type developer and still retain satisfactory sound quality. It 
was found, however, that the negative density tolerances were very 
narrow. While it is possible to develop both negative and print in 
the print developer and obtain good quality, it is not advisable to do 
so in production, as all phases of recording and processing would 





have to be controlled within impractical limits to obtain consistently 
good results. 

Referring again to Fig. 1, this film exposed with ultraviolet light 
has a higher density speed than the 1302 exposed with unfiltered light. 
A 1372 negative was also made 
with unfiltered light and de- 
veloped in the negative de- 
veloper. It compares favorably 
with the 1372 exposed with 
ultraviolet light and has the 
advantage of saving approxi- 
mately 0.7 amp in lamp cur- 

It was considered desirable 
to investigate the combination 
of the 2 previous tests, that is, 
exposing the 1372 film with 
unfiltered light and develop- 
ing the negative in the print- 
type developer. Prints from 
this negative show the same 
narrow tolerances found on 
the prints from the 1372 
exposed with ultraviolet light 
and developed in the print- 







FIG. 10. Processing tolerances, 1302. 
Negative exposed with unfiltered light. 
Negative and prints developed in vari- 
able-area negative developer. 

type developer. 

The writer wishes to take 
this opportunity to express her appreciation for the generous help 
provided by R. V. McKie. 


1 BAKER, J. O., AND ROBINSON, D. H.: "Modulated High-Frequency Record- 
ing as a Means of Determining Conditions for Optimal Processing," /. Soc. Mot. 
Pict. Eng., XXX, 1 (Jan., 1938), p. 3. 



Summary. The brightness, or candlepower per unit area, of a light source is 
an important determinant of the intensity in a projected light beam. A method is 
described for measuring brightness of carbon arcs by determining the light intensity 
on a projected image of the source. The principles of the system and the equipment 
used for the measurement are discussed. 

The significance of source brightness, or candlepower per unit area, 
in light projection systems such as are used in the motion picture 
industry has been discussed in numerous instances. 1 Briefly, the 
light intensity on the motion picture screen increases with an increase 
in source brightness if all other factors remain constant. 

The purpose of this paper is to describe the technique used in our 
laboratories to measure the brightness of carbon arc sources. Basi- 
cally the method involves the measurement of the light intensity on 
a projected image of the source. 

Fundamental Considerations. In Fig. 1 is shown a sketch in 
which the letter B represents a point on the crater of a positive car- 
bon with a brightness B expressed in candlepower per unit area, 
and the letter E refers to the light intensity at the corresponding 
point on the image, expressed in lumens incident per unit area. 
The angle a indicated in the sketch is the half angle of the cone of light 
received through the lens from B, while is the half angle of the cone 
of light collected by the lens from B. 

The mathematical expression 

E = irB Sin 2 a (1) 

relates the light intensity E to the brightness B, the constant TT, and 
the angle a. This equation holds when the source obeys Lambert's 
cosine law of emission and when the lens used to form the image ful- 

* Presented Oct. 16, 1944, at the Technical Conference in New York. 
** National Carbon Company, Inc., Fostoria, Ohio. 



fills the sine condition. Since these fundamental relationships and 
conditions have been thoroughly explained elsewhere, 2 no further 
elaboration will be given to them here. 

According to the laws of optics, 2 the magnification of the crater at 
the image is given as 

M = 

sin e 

sin a 

where M signifies linear magnification. If 6 and a are small enough 
then for all practical purposes 



FIG. 1. Sketch of optical system for brightness determination. 

6 may also be expressed as 

sin B = 6 
and sin a = a 

6 = D/2R 

where D is the diameter of the lens stop and R is the distance from the 
lens stop to the crater as illustrated in Fig. 1 . Therefore the funda- 
mental expression 

E = irB sin 2 a 

may be rewritten as 

B = EM 2 X 

4R 2 


Eq (2) does not take into consideration lens transmission which is 
neglected in the basic relationship expressed by Eq (1). If is meas- 
ured in foot-candles and distances are expressed in feet, then Eq (2) will 
result in a value of B in candlepower per sq ft. In order to include 
the lens transmission factor and to express B in the more generally 



Vol 45, No. 1 

used term of candlepower per sq millimeter, the following expression 
results from Eq (2) : 


5 = X ^ X i* 2 X 

T * *D* * (12 X 25.4) 2 


where 5 = candlepower per sq millimeter of source. 
E foot-candles intensity on image. 
M = linear magnification. 
T = lens transmission. 
R = distance lens stop to crater. 
D = diameter of stop in contact with lens. 

(12 X 25 41 2 ~ constant to convert brightness from candlepower per sq ft 
to candlepower per sq millimeter. 

FIG. 2. View of photocell, recorder, and mechanism for measuring 

It can be seen from Eq (3) that if the quantities M the magnifica- 
tion, T lens transmission, R the crater to lens distance, and D 
the diameter of the lens stop, are known, the determination of bright- 
ness resolves itself into a measurement of light intensity in foot- 
candles on the projected image of the positive carbon crater. 

Method and Equipment for Brightness Measurement. The 
fundamental optical system used in measuring brightness is as illus- 



trated in Fig. 1. The carbon arc sources and the optical system ful- 
fill Lambert's law and the sine condition sufficiently well over the 
2 or 3 degrees total collecting angle involved to meet the fundamen- 
tal requirements of Eq (I). 

The light intensity in foot-candles on the image is measured with a 
photocell having a spectral response approximating the eye sensitivity 
curve. A small lens of approximately 7 in. focal length is used, to 
project the crater image. The magnification is calculated from the 

FIG. 3. Close-up of motor and driving mechanism for photocell. 

ratio of the image and object distances, and is also verified by pro- 
jecting the image of an accurately gauged piece of metal. The lens 
transmission is determined and approximates 0.92. The lens to 
crater distance and the diameter of the stop in contact with the 
lens are also measured. All these determinations result in the speci- 
fication of E, M, T, R, and D in Eq (3). 

Fig. 2 shows the complete assembly developed by our laboratory 
for measuring brightness. This includes photocell, recording meter, 
and mechanical arrangements. 

It has been found convenient to use a magnification of 25.4: 1 since 



Vol 45, No. 1 

this makes one inch at the image equivalent to one millimeter at the 
source. The photocell is masked to a Vz-m. diameter which corre- 
sponds to V 2 -mm diameter at the crater. The photocell output is 
recorded with a microammeter calibrated in foot-candles, and a value 
of D, the diameter of the lens stop, is so chosen that the photocell out- 
put recorded in microamperes can be multiplied by a round number 
such as 100 or 200 to read directly in candlepower per sq millimeter. 
Suitable corrections are made for any departure from linearity in the 
foot-candle versus microammeter calibration. 

Although the fundamental Eq (1) upon which the method of measur- 
ing brightness depends is defined only for points on the axis of the 
system, the angles involved in measuring brightness for off-axial 

FIG. 4. Record of brightness distribution across positive carbon crater. 

positions are so small that corrections can be neglected. Accordingly 
the distribution of brightness accross the crater is determined by 
passing the photocell through the crater image. The photocell is 
carried across the projected image by a synchronous motor-driven 
screw. Fig. 3 shows the arrangement of the motor and driving 
mechanism. The output of the photocell is recorded on a chart 
driven by another synchronous motor. 

The brightness measuring unit has a number of convenient features. 
The image board on which the photocell and driving mechanism are 
mounted can be rotated through any angle between and 90 degrees 
so that the crater image can be traversed by the photocell to give a 
brightness distribution across the crater in a horizontal, vertical, or 
any intermediate direction. The frame on which the image board is 
mounted is on a carriage to permit easy movement and also to facili- 
tate horizontal alignment of the image on the board. A crank and 


screw arrangement makes it possible to adjust the vertical alignment 
of the image on the board. The motor driving the photocell is revers- 
ible and by stopping the motor the photocell can be used to record 
brightness versus time at any part of the crater image. The photo- 
cell can be disengaged from the screw drive very quickly by means of 
a simple clutch to permit rapid positioning of the photocell on the 
image board. 

Since the photocell is carried across the image by a synchronous 
motor-driven screw and the recording microammeter chart is driven 
by a synchronous motor, various positions on the chart are directly 
related and measurable in terms of position on the source. A chart 
of crater brightness distribution as recorded by this apparatus is illus- 
trated in Fig. 4. Since the height of the curve is representative of 
candlepower per sq millimeter and the chart travel is a measure 
of crater dimensions, it is possible to use such curves to determine 
the crater candlepower by graphical integration. 

The mechanism and method described have been valuable, depend- 
able, and rapid for making the important measurement of brightness. 
The fundamentals of the method are based on well-known and well- 
described principles. The mechanism employed to apply these prin- 
ciples was developed to provide a laboratory tool simple and conveni- 
ent to use. 


1 COOK, A. A.: "A Review of Projector and Screen Characteristics and Their 
Effects Upon Screen Brightness," /. Soc. Mot. Pict. Eng., XXVI, 5 (May, 1936), 
p. 522. 

2 HARDY, A. C., AND PERRIN, F. H.: "The Principles of Optics," McGraw- 
Hill Book Co., Inc. (New York), 1932, ch. XIX, pp. 409^16. 


Ed. Note. On May 25, 1945, the Federal Communications Commission issued an 
official report allocating frequencies to the various nongovernmental radio services 
from 25,000 to 44,000 and from 108,000 to 30,000,000 kilocycles. Subsequently the 
FCC also issued a final report on allocations to the various nongovernmental radio 
services from 44,000 to 108,000 kilocycles. 

The Society of Motion Picture Engineers, through its representative, Paul J . 
Larsen, submitted recommendations for allocations of frequencies for a national 
theater television service at different hearings held before the FCC, which have been 
reported in previous issues of the Journal. \44, 2 (Feb., 1945), pp. 105-137; 44, 4 
(Apr., 1945), pp. 263-274.} 

The FCC granted allocations of frequencies for experimentation of theater television 
on a "parity of opportunity" basis with television broadcasting. The following 
tabulation compares the SMPF, requests with the proposed and final allocations 
granted by the Commission: 


vSMPE Request 
Band Total 

From To MC 

FCC Proposed. Jan. 15, 1945 FCC Granted May 25, 1945 

Band Total Band Total 

From To MC From To MC 































Total below 10,000 








Total above 1000 



Total Megacycles 1500 11.290 11,490 

* Experimental to be discontinued when needed for broadcast service. 

The allocations of frequencies for experimental theater television granted by the 
FCC now permit the motion picture industry to proceed with plans for this new 
service by filing of applications in specific portions of the frequency spectrum. 


Excerpts from the final FCC report are given here to complete the published record 
of SMPE activities in this matter. 



TO 30,000,000 KILOCYCLES* (DOCKET NO. 6651) 


This report contains the Federal Communications Commission's allocations to 
the various non-governmental radio services from 25,000 to 44,000 and 108,000 to 
30,000,000 kilocycles. No final allocations are being made at this time for the 
portion of the spectrum between 44 and 108 megacycles. Instead three different 
alternative allocations are set forth... and further measurements and tests will 
be conducted during the summer in order to aid the Commission in deciding the 
best allocation for this portion of the spectrum. 

This report should be read in conjunction with the proposed report of alloca- 
tions above 25,000 kilocycles, dated January 15, 1945. That proposed report is 
adopted as the final report of the Commission except as modified or brought up 
to date by this report. 

It is contemplated in the near future to issue a proposed report for allocations 
below 25,000 kilocycles and as soon thereafter as possible a final report for such 
allocations. It is the Commission's intention thereafter to issue one final report 
bringing together in one place all of the material in the various proposed reports 
and final reports. 





The Commission's proposed report for allocations from 25,000 kilocycles to 
30,000,000 kilocycles was dated January 15, 1945. Provision was made at that 
time for oral argument on this report by any interested person. By subsequent 
public notice** it was provided that in addition to presenting oral argument inter- 
ested persons would be permitted to present any additional evidence or material 
that might have been developed since the close of the hearing. These further 
hearings were held on February 28 to March 3, 1945, inclusive, and March 12, 
1945. A list of the persons who appeared and presented statements and testi- 
mony at such oral argument is included in the appendix. Also included is a list 
and description of the exhibits that were introduced at these further hearings. 

It appeared at the further hearing that much of the data and evidence con- 
cerning propagation was of a classified nature and hence could not be discussed in 

* Dated May 25, 1945. 

Copies of all public notices issued subsequent to the proposed report are in- 
cluded in the appendix. (Not published with these excerpts Ed.) 

* * 


an open hearing. It was also apparent that it was not possible to reach a final 
determination unless this evidence were produced and made subject to cross- 
examination by interested persons. Accordingly, a closed hearing was arranged 
under the supervision of the military services for the presentation of such material. 
These hearings were held on March 12 and 13, 1945. In addition to members of 
the Commission's staff, engineers representing interested persons who were 
cleared for that purpose were permitted to attend the hearing and to cross- 
examine the witnesses. It is the Commission's hope that as soon as security 
regulations permit, much or all of this classified material will be declassified and 
made available to the public. But until that happens, the material cannot be 
divulged. In this report, however, the Commission has relied upon the material 
presented in the closed hearings where relevant although for security reasons de- 
tailed discussion of this material is not possible at this time. 


One brief, that of Columbia Broadcasting System, Inc., objected to the alloca- 
tions proposed to be made to theater television. Oral argument was presented by 
Mr. Joseph H. Ream on behalf of that organization (Tr. 4910) and by Mr. Paul J. 
Larsen on behalf of the Society of Motion Picture Engineers (Tr. 4979-4986). 

The Commission's Report (pp. 188-189) stated that while no specific frequen- 
cies would be allocated to theater television, consideration would nevertheless be 
given to applications for experimental authorizations in this field on frequencies 
between 480 and 920 megacycles allocated to broadcasting on the basis that the 
use of these frequencies would be discontinued when needed for the broadcast 
service. In addition, the several bands of frequencies above 1900 megacycles 
allocated for general experimentation were made available for experimentation 
with intra- and inter-city relay of theater television programs. 

The Columbia brief objected to the assignment of any frequencies to either sub- 
scription radio or theater television. The reasons for denying these services any 
frequencies were presented in detail as to subscription radio and were stated to 
"apply with at least equal force to the requested use of frequencies for theater 
television" (Brief, p. 36), although not separately treated. In brief, Columbia's 
position is that theater television is not broadcasting and hence should be assigned 
frequencies allocated to point-to-point services, if at all. In addition, Columbia 
made the point that theater television could use wire lines. 

Mr. Larsen disputed the validity of these reasons, as applied to theater tele- 
vision, stating that theater television was admittedly not to be considered a 
broadcast service, and the Society's request for frequencies had been made on the 
express basis that the Commission would classify theater television as communica- 
tions of a private nature so as to differentiate the service from broadcasting. 
With respect to utilization of wire lines for theater television he stated that wire 
lines are technically and economically impractical as well as inflexible and that 
present wire lines are not suitable for transmissions of channel widths greater than 
four megacycles. He pointed out that 85,000,000 people per week are enter- 
tained at theaters and, therefore, the proposed theater television service is 
obviously not of a limited character. 

The Society of Motion Picture Engineers accordingly requested the Commission 
to reaffirm its experimental allocations for the theater television service in the 


several experimental bands above 1000 megacycles with the single modification 
that the final allocation make it clear that multiple address stations would be 
permissible at these frequencies as well as in the 480-920 megacycle band. It also 
requested the Commission to reaffirm its allocation of frequencies between 480 
and 920 megacycles, but with the following two modifications: (a) Place tele- 
vision on a "parity of opportunity basis" with television broadcasting. (It was 
explained that this did not mean an equal sharing of these frequencies as theater 
television, employing directive networks, will require far less frequencies than 
broadcasting; but theater television felt that its service had, with television 
broadcasting, an equal responsibility to the public in the visual and aural enter- 
tainment field and allocations should accordingly be made to theater television as 
a competitive service to television broadcasting.) (b) Make the allocations to 
theater television between 480 and 920 megacycles permanent if the results of 
experimentation demonstrate that theater television can make better use of this 
band than of higher frequencies. 

Both these requests the Commission has decided to grant, but not with the 
modifications suggested with respect to the use of the 480-920 megacycle band. 
Despite the probable future importance of the theater television service, it is felt 
that insufficient information is now available to make a specific determination of 
the needs of this service for spectrum space. Provision will therefore be made for 
experimentation in this new service in order to obtain information for a more pre- 
cise determination of need at a later time when further consideration will be given 
to the availability of wire circuits for this type of service. It is expected, more- 
over, that the band 480-920 megacycles will be used primarily for television 
broadcasting to the public, with higher frequencies being more properly utilized 
by theater television and relay operation. Theater television experimental 
operation permitted in the 480-920 megacycle band will therefore be subject to 
the use of and need for this band for television broadcast service directly to the 

The allocation of frequencies as proposed in the Commission's Report of 
January 15, 1945, is accordingly made final except that the bands of experimental 
frequencies above 1000 megacycles allocated to Non-Government, Fixed and 
Mobile service and available for theater television experimental use, including 
multiple address purposes if the need for such use can be established, will be, as 
revised by this Report, as follows : 1325-1375, 1750-2100, 2450-2700, 3900-4400, 
5650-7050, 10,500-13,000, 16,000-18,000 and 26,000-30,000 megacycles. 



Summary. The use of a polar o graph in the analysis of photographic fixing 
baths for aluminum alum, chrome alum, sulfite, and silver is described. A Fisher 
Elecdropode was used in this work. The polar o graphic method is shown to have an 
accuracy within 5 per cent for each of these 4 materials, which is satisfactory for 
photographic purposes. The reproducibility of the method is shown to be within 3 
per cent. This method of analysis for aluminum alum, chrome alum, and silver is 
in use in this laboratory instead of the gravimetric method which is considered more 
cumbersome and time consuming. The polarographic method of analysis for sulfite 
has also been found more convenient in this laboratory than the previously used volu- 
metric method. 


During the 24 years of its existence, the polarographic method of 
analysis has developed rapidly and many articles have been published 
describing its use in the analysis of various materials, among which 
were those by Evans, Hanson and Glasoe 1 - 2> 3 concerning the meas- 
uring of the copper, sulfide, elon, and iodide contents of developers. 
As a result of the successful application of the polarographic method 
of analysis to developers, this laboratory decided to purchase a 
polarograph to be used in chemical technical service to the motion 
picture industry. Upon obtaining this instrument, it was success- 
fully employed in an analysis of a developer for copper contamination 
and in a survey of the water from various Hollywood motion picture 

While the polarograph has been used satisfactorily for many 
things, as far as the authors could ascertain, no work has been re- 
ported on the application of the polarographic method of analysis to 
fixing baths which this laboratory is requested frequently to analyze. 
The method of analysis for alum, sulfite, and silver used has been that 
reported by Atkinson and Shaner. 4 While this method is satisfactory 

* Presented Oct. 17, 1944, at the Technical Conference in New York. 
** Motion Picture Film Dept., Eastman Kodak Co., Hollywood. 


from the standpoint of accuracy, it is rather cumbersome and time 
consuming. It was decided to investigate the application of the 
polarograph to fixing bath analysis, with the hope that a less time- 
consuming method could be evolved. 


The instrument used, the Fisher Elecdropode, is a manually oper- 
ated instrument for obtaining polarographic waves. A water bath 
was used to maintain a constant temperature of 25 C. The procedure 
followed was that described by Lingane, 5 in which the Ilkovic equa- 

i d = kn D l /*Cm*/tf/6 (1) 

in which i d is the diffusion current in microamperes, n is the number 
of electron equivalents per molar unit of electrode reaction, D is the 
diffusion coefficient (sq cm per sec) of the reducible or oxidizable 
substance, C is its concentration in millimoles per liter, m is the rate 
of mercury flow from the dropping electrode in mg per sec, and t is 
the drop time in seconds, is used. According to Lingane, 5 k, n, and 
D are independent of the characteristics of the dropping mercury 
electrode capillary and the quantity kn D l/ \ or /, is experimentally 
determinable as i d /(Cni /3 t l/6 ) I is referred to as the "diffusion cur- 
rent constant." If / is substituted in the Ilkovic Eq (1) and then it 
is solved for C, the equation, 

C ' ' 

is obtained. Since m is independent of the potential of the dropping 
electrode and the medium in which the drops form, 5 it need not be 
measured more frequently than once a week, unless there is reason to 
believe that the capillary has become contaminated. However, t 
varies with the medium and with potential 5 so it is necessary to 
measure / at the potential at which the diffusion current is measured 
every time a determination is made. Since the diffusion current 
varies with the temperature, 6 it is necessary to control the tempera- 
ture within i/ 2 C. 

The current scale on a Fisher Elecdropode is not calibrated in 
amperes, so in order to use the equation given above, it was necessary 
to calibrate the galvanometer. The method used was that described 
by Kolthoff and Lingane 7 in which an 9999-ohm resistance box was 
hooked into the circuit in place of the electrolysis cell and calculations 

22 V. C. SHANER AND M. R. SPARKS Vol 45, No. 1 

using Ohm's law were made. Diffusion current constants were 
calculated by measuring i d for varying concentrations of the material 
in question when m and t were known and using Eq (2). The con- 
stants thus obtained are given in Table 1. Before each determina- 
tion, nitrogen was bubbled through the solution for 5-10 min to free it 
of dissolved gases. 


Diffusion Current Constants 































Aluminum. A polarographic exploration of a fixing bath re- 
vealed a large wave which completely obscured the aluminum wave, 
so it was necessary to precipitate the aluminum from the fixing bath. 
A 50-ml sample of the fixing bath was taken and made just alkaline 
with ammonium hydroxide, heated to boiling, and filtered. The 
white precipitate of aluminum hydroxide was washed on the filter 
paper with hot ammonium nitrate and washed into a beaker with 
distilled water. The mixture was made acid with acetic acid and 
heated until the aluminum hydroxide dissolved. The volume was 
made up to 250 ml. It was discovered that the aluminum wave was 
covered by the hydrogen ion wave from the acid. For this reason, 

the following indirect method of determining aluminum was devised. 


The hydrogen of the OH group of 8-hydroxyquinoline, 

is replaceable by aluminum in dilute acetic acid solution which is 
buffered with ammonium acetate. 8 8-Hydroxyquinoline is polaro- 
graphically determinable in a base electrolyte of ammonium acetate. 
Therefore, if the aluminum hydroxide precipitate from a fixing bath is 
dissolved in dilute acetic acid and allowed to react with an excess 
amount of 8-hydroxyquinoline reagent of known composition and a 
sample of the supernatant 8-hydroxyquinoline analyzed polaro- 
graphically, the aluminum concentration can be calculated from the 
change in the 8-hydroxyquinoline concentration. 

July, 1945 



Take a 50-ml sample of fixing bath, make alkaline with ammonium hydroxide, 
filter, and wash with hot ammonium nitrate solution. Dissolve the precipitate in 
very dilute acetic acid solution and dilute to 250 ml with distilled water. Heat 
10 ml of this solution to about 70 C and add 10 ml of 8-hydroxyquinoline reagent 
(25 gm 8-hydroxyquinoline in 60 ml glacial acetic acid, diluted to 2 liters with cold 
water) and 25 ml of 2 TV ammonium acetate solution. Dilute to 100 ml and let 
stand until the precipitate settles. Add one ml of the supernatant liquid to 5 ml 
of 2 N ammonium acetate and 3 drops of 6 per cent gelatin. Dilute to 20 ml and 
determine polarographically the 8-hydroxyquinoline concentration. 



I 00 





0-.2 -4 -.6 -.8 -1.0 -1.2 -1.4 -1.6 -1.8 

FIG. 1. 8-Hydroxyquinoline. 

The difference in the value obtained and the 8-hydroxyquinoline 
concentration in the reagent diluted as described is the amount which 
reacted with the aluminum. A typical 8-hydroxyquinoline curve is 
shown in Fig. 1. The wave starts at a potential of about 0.8 v and 
levels at a potential of about 1.2 v. Typical diffusion current con- 
stants are shown in Table 1 . 

Sample Calculation. If i d = 2.197 /za, m = 2.2019 mg per sec, t = 2.68 sec, and 



Vol 45, No. 1 

/ = 3.30, the concentration of 8-hydroxyquinoline in the electrolysis cell may be 
calculated by substitution in Eq (2) as follows : 



(2.2019) 2 / 3 (2.68) 1 / 6 3.30 

= 0.337 millimole of 8-hydroxyquinoline per liter. 
Since one ml of unknown is diluted to 20 ml in the electrolysis cell, 




< 100 


^ 80 



-.2 -4 & -.8 -1.0-12 -14-1.6 -1.8 

FIG. 2. Chromium. 

0.3337 X 20 

= 0.006674 molar 8-hydroxyquinoline in solution with 

aluminum precipitate. 8-Hydroxyquinoline reagent is 0.0858 M, but is diluted 
10 times before it is allowed to react with aluminum. 

0.00858 - 0.00667 = 0.00191 N aluminum 
In order to convert this value to gm per liter of aluminum alum 

0.00191 X 100 X 250 X 474.38 

= 15.10 gm potassium aluminum per 
3 X 50 X 10 

liter of fixing bath. 


Chromium. As in the case of the aluminum wave, the chromium 
wave was found to be obscured by the fixing bath wave. Chro- 
mium was precipitated and washed with ammonium nitrate like 
the aluminum. The blue-green precipitate of chromium hydroxide 
was dissolved in dilute hydrochloric acid and the volume of solution 
made up to 500 ml. One ml of this solution was taken for polaro- 
graphic analysis, added to a base electrolyte of 5 ml of 0.4 N calcium 
chloride, and diluted to 20 ml with distilled water. No gelatin was 
added. The chromium wave starts at a potential of about 0.6 v 
and levels off at a potential of about 1.2 v. 9 A typical wave is 
shown in Fig. 2 and typical diffusion current constants for chromium 
are shown in Table 1. 

Sample Calculation. If i d = 1.0 /xa, m = 2.2019 mg per sec, t = 3.18 sec, and 
/ = 1.58, then the concentration of chromium in the electrolysis cell may be calcu- 
lated by substitution in Eq (2) as follows: 


(2.2019) 2 / 3 (3.18) 1/6 1.58 
= 0.3084 millimole Cr +++ per liter. 

Then to convert this value to gm per liter of chrome alum 
0.3084 X 20 X 500 X 499.42 

1000 X 50 

30.80 gm chrome alum per liter of fixing 

Sulfite. In neutral or alkaline solution, sulfite ion is not reducible 
at the dropping mercury electrode, but in acid solutions a well- 
defined wave is obtained. 10 The diffusion current will vary with the 
/>H of the solution, so it is necessary to use a buffer as the base elec- 
trolyte. A pH. 4.0 buffer composed of 27.48 gm per liter of Na2H 
PCV12 H 2 O and 5.90 gm per liter of citric acid proved satisfactory. 
It was noticed that if the diffusion current of the sulfite ion was 
greater than 15 jua, the current was not directly proportional to the 
concentration of the sulfite ion present. Therefore, the size of the 
sample of fixing bath must be chosen accordingly. If the fixing bath 
contains less than 10 gm per liter, one ml sample may be used; if it 
contains more than 10 gm per liter, or gives a diffusion current 
greater than 15 //a, a 1 /2-ml sample may be used. 

A 1 /2-ml sample of fixing bath may be obtained by diluting 25 ml of 
the fixing bath to 50 ml with distilled water and taking one ml of the 
diluted mixture. The fixing bath sample is added to 10 ml of the pH 



Vol 45, No. 1 

4.0 buffer and 3 drops of 6 per cent gelatin, and the whole is diluted to 
20 ml for polaro graphic measurement. A typical wave is showri in 
Fig. 3. The wave starts at a potential of about 0.2 v and levels off 
at a potential of about 0.6 v. Typical diffusion current constants 
are shown in Table 1 . 


S 500 


LJ 400 

5 300 




D 100 


-.2 -4 -.6 -.8 -I.0-l.2-l.4-l.6-l.8-a0 

FIG. 3. Sulfite. 

Sample Calculation. -If i d = 5.9 /*a, m = 2.2019 mg per sec, * = 3.04 sec, and 
7 = 1.35, the concentration of sulfite in the electrolysis cell can be expressed by 
substitution in Eq (2) as follows : 



(2.2019) 2 / 3 (3.04) 1 / 6 1.35 
= 2.146 millimoles SOs*" per liter of solution. 
In order to convert this value to gm per liter of Na 2 SO 3 
2.146 X 20 X 126.05 


5.41 gm sodium sulfite per liter of a fixing bath. 

July, 1945 



Silver. A well-defined diffusion current is obtained from the 
reduction of silver thiosulfate ion from a solution of sodium thio- 
sulfate. 11 This wave starts from applied potential. To assure a 
large concentration of thiosulfate ions, 10 ml of 0.1 TV sodium thio- 
sulfate were used as the base electrolyte. One ml of the unknown 
fixing bath and 3 drops of 6 per cent gelatin to prevent maxima were 
added. This solution was diluted to 20 ml and placed in the elec- 



+.8 -.6 +4 t2 -2 -4 -.6 -.8 -1.0-1.2 

FIG. 4. Silver. 

trolysis cell for polarographic determination. A typical wave is 
shown in Fig. 4. It is advisable to make a determination on a blank 
solution, so that any current carried by the base solution alone may 
be subtracted from the current carried by the silver and the base 
solution. To determine the diffusion current constant /, known 
quantities of silver nitrate were added to a fixing bath and the diffu- 
sion currents were measured as described above. The diffusion cur- 
rent constants thus found are reported in Table 1. The average 
value was used in analysis. 

28 V. C. SHANER AND M. R. SPARKS Vol 45, No. l 

Sample Calculation. If the measured value of i d = 4.9 /^a, m = 2.2019 mg per 
sec, t = 3.22 sec, and I = 1.58, then the concentration of silver in the electrolysis 
cell may be determined by substitution in Eq (2) as follows: 


(2.2019) 2/s (3.22) 1 / 6 1.58 
= 1 .508 millimoles Ag + per liter or 
1.508 X 20 X 107.88 


= 3.25 gm silver per liter. 

Hypo. An attempt was made to analyze a fixing bath for sodium 
thiosulfate by polarographic means. The thiosulfate ion gives a 
polarographic anodic wave at a potential of about 0.25 v. This 
value is shifted to a more positive potential if the thiosulfate ion con- 
centration is very great. 1 ' 2 In order to put the thiosulfate ion con- 
centration of a fixing bath within the range determinable on a polaro- 
graph, the fixing bath would have to be diluted. This dilution fre- 
quently resulted in decomposition of some of the thiosulfate into 
sulfur and sulfite. For this reason, it is recommended that the iodine 
titration method of Atkinson and Shaner 4 be used to determine the 
total amount of hypo and sulfite present. The sulfite can be deter- 
mined polarographically, subtracted from the total, thus giving the 
quantity of hypo present without employing a formaldehyde titration. 


Accuracy. To determine the accuracy of the method, 4 analyses 
of fixing baths of known composition were made. The results of 
these analyses are given in Table 2. From these data it can be seen 
that the method is accurate within 5 per cent for aluminum, chro- 
mium, and silver. The high per cent error of sodium sulfite in the 
chrome alum fixing bath is believed to be caused by the escape of part 
of the sulfite added to the fixing bath as sulfur dioxide, because of the 
high degree of acidity of a chrome alum fixing bath. The sulfite con- 
centration of the fixing bath was checked iodometrically and found to 
agree with the polarographic results. It is to be noted that this high 
error did not occur in the analysis of potassium aluminum alum fixing 
baths. For these reasons and because of the reproducibility of a 
sulfite constant, it is concluded that the accuracy of the polaro- 
graphic sulfite analysis is within 5 per cent. Since it is a generally 
accepted fact that a change in concentration of 10 per cent of a 
chemical in a fixing bath is necessary to show any photographic effect, 

July, 1945 



the accuracy of this analytical 
method is quite satisfactory. 

Reproducibility. The re- 
producibility of the method 
was tested by analyzing a 
potassium aluminum alum 
fixing bath and a potassium 
chromium alum fixing bath 
6 times apiece. The data 
thus collected are given in 
Tables 3 and 4. From Table 
3, it can be seen that in (> 
analyses of a potassium alu- 
minum alum fixing bath, the 
greatest deviation from the 
mean in a sulfite analysis is 
1.86 per cent, in an aluminum 
analysis the greatest deviation 
from the mean is 2.19 per 
cent, and in a silver analysis 
the greatest deviation is 2.(> 
per cent. From Table 4 it can 
be seen that in 6 analyses of 
a chrome alum fixing bath the 
greatest deviation from the 
mean in the sulfite analysis is 
1.40 per cent, the greatest de- 
viation from the mean in the 
chromium analysis is 2.22 per 
cent, and the greatest devia- 
tion from the mean in the 
silver analysis is 2.83 per cent. 
Thus, it may be concluded 
that the reproducibility of 
analyses made by this method 
is quite satisfactory. 

Time of Analysis. The po- 
larographic method of analy- 
sis is much less time consum- 
ing than the regular method 

CO O Oi 

co -f i-I 




B? o : o 





CO -f O 

i-t iO 00 
00 00 i-H 

O (M O 

CO 1-1 

r O ~ 

& s t ^ ^ 

*o w ^ co o 


o o o 

- ** <M 

3 .2 

s s 

^ ^ ^ 

< O 

I I 5 

't/3 'en 

t/3 CO M 

of 9 

n ? O 

OH fXi C/3 


o 3 


c c 

OJ <U 

S J 

S E 
o o 




Vol 45, No. 1 

of analysis. A polarographic silver analysis requires only about 15 
min as compared with the considerably longer time required for a 
gravimetric silver analysis. Polarographic chromium and aluminum 
analyses require the same length of time for precipitation as the 
gravimetric chromium and aluminum analyses, but do not require 


Six Analyses of Potassium Aluminum Alum Fixing Bath M 

Sodium Sulfite 

Potassium Aluminum Alum 

Per Cent 

Per Cent 




from Mean 


from Mean 


+ 1.02 




- 1 . 86 




+ 1.79 






+ 1.66 




+ 1.34 




+ 1.28 

Average : 






Per Cent 
Gm/L from Mean 




Six Analyses of Chrome Alum Fixing Bath N 

Sodium Sulfite 

Per Cent 



from Mean 





+ 1.40 







Average : 





Per Cent 
from Mean 


+ 1.41 













Per Cent 
from Mean 














the long time for ignition and weighing of the precipitates. Polaro- 
graphic sulfite analysis requires as much time as the iodometric 
method, but the use of formaldehyde and ice, which is required in 
the iodometric method, 4 is not necessary. 

It should be emphasized that the use of a polarograph, like any 


scientific precision instrument, requires considerable experience by a 
qualified chemist before reliable results can be obtained. The 
authors have found the text "Polarography" by Kolthoff and Lingane 
to be almost indispensable as a guide and reference in developing a 
technique for the operation of this instrument. 

The authors wish to express their grateful appreciation to Emery 
Huse under whose guidance this work was undertaken and completed. 
In addition, they wish to acknowledge the helpful suggestions of 
various members of the Kodak Research Laboratories. 


(1) Polarographic methods of analysis may be satisfactorily ap- 
plied to the aluminum alum, chrome alum, sulfite, and silver contents 
of photographic fixing baths. 

(2} The accuracy of the polarographic analytical method is within 
5 per cent for these 4 photographic fixing bath constituents. 

(3) The results of the polarographic analyses for the above-named 
fixing bath constituents are reproducible within 3 per cent. 

(4) It was the experience of the authors that the polarographic 
method of analysis for sodium thiosulfate in fixing baths is less 
practical than the conventional iodometric method of analysis. 

(5) Analysis of fixing baths for aluminum alum, chrome alum, and 
silver by polarographic means is less cumbersome and time consuming 
than analysis by gravimetric methods. The polarographic method 
of analysis for sulfite has also been found more convenient in this 
laboratory than the previously used volumetric method. 


1 EVANS, R. M., HANSON, W. T., AND GLASOE, P. K.: "Iodide Analysis in 
M. Q. Developers," J. Soc. Mot. Pict. Eng., XXXVIII, 2 (Feb., 1942), p. 180. 

2 EVANS, R. M., HANSON, W. T., AND GLASOE, P. K.: "Synthetic Aged De- 
velopers by Analysis," /. Soc. Mot. Pict. Eng., XXXVIII, 2 (Feb., 1942), p. 188. 

3 EVANS, R. M., HANSON, W. T., AND GLASOE, P. K. : "Copper and Sulfide in 
Developers," J. Soc. Mot. Pict. Eng., XL, 2 (Feb., 1943), p. 88. 

4 ATKINSON, R. B., AND SHANER, V. C. : "Chemical Analysis of Photographic 
Developers and Fixing Baths," /. Soc. Mot. Pict. Eng., XXXTV, 5 (May, 1940), p. 

5 LINGANE, J. J.: "Systematic Polarographic Metal Analysis," Ind. Eng. 
Chem., Anal. Ed., 15 (1943), p. 588. 

6 KOLTHOFF, I. M., AND LINGANE, J. J.: "Polarography," Interscience Pub- 
lishers, Inc. (New York), 1941, p. 249. 

7 KOLTHOFF, I. M., AND LINGANE, J. J. : Ibid., p. 228. 


8 TREADWELL, F. P., AND HALL, W. T. : "Analytical Chemistry," John Wiley 
and Sons, Inc. (New York), 1935, p. 98. 

9 KOLTHOFF, I. M., AND LlNGANE, J. J. : Ibid., p. 291. 

10 KOLTHOFF, I. M., AND LINGANE, J. J. : Ibid., p. 312. 

11 KOLTHOFF, I. M., AND LINGANE, J. J.: Ibid., p. 259. 

12 KOLTHOFF, I. M., AND LINGANE, J. J.: Ibid., p. 326. 



J. W. McNAIR' 

This is the third report on the work of the American Standards 
Association War Committee on Photography and Cinematography, 
Z52, which has been presented before a semi-annual technical con- 
ference of the Society. 

Since the last conference some 16 new American War Standards 
have been completed, approved, and made available for the use of the 
Armed Forces and industry. This brings to 41 the total number of 
American War Standards for photography and motion pictures thus 
far completed. Active work on 30 additional projects is well under 
way while work authorized on 7 other subjects will be begun as soon 
as the time and personnel are available. 

The standards which have been completed since the October, 1944, 
Conference of the Society include several which will be of considerable 
future importance to the science of motion picture engineering. I 
have been told by a number of members of the Society that, in their 
opinion, some of these can be adopted with little, if any, change as 
regular American Standards under the Society's sponsorship of the 
ASA Sectional Committee on Motion Pictures, Z22. 

Most important of these new standards to be completed since last 
October is the American War Standard Nomenclature for Motion 
Picture Film Used in Studios and Processing Laboratories, Z52.14- 
1944. Some 5000 copies of this standard have been required by the 
Armed Forces alone for distribution to the service personnel con- 
cerned, while an additional 3000 copies have been made available to 
the industry in pamphlet form. 

Moreover, the Society has deemed this War Standard of such im- 

* Presented May 14, 1945, at the Technical Conference in Holly-wood. 
"* Secretary, War Committee on Photography and Cinematography, Ameri- 
can Standards Association, New York. 


34 J. W. McNAIR Vol 45, No. 1 

portance in clarifying the confusion in nomenclature which exists 
that it has reprinted the document in toto in the April, 1945, issue of 

With regard to the quality control of 16-mm film, 3 additional 
standards have been completed : the American War Standard Method 
of Making Intermodulation Tests on Variable- Density 16-Mm 
Sound Motion Picture Prints, Z52. 15-1944, American War Standard 
Method of Making Cross-Modulation Tests on Variable- Area 16-Mm 
Sound Motion Picture Prints, Z52.39-1944, and American War 
Standard Specification for Warble Test Film Used for Testing 16- 
Mm Sound Motion Picture Equipment, Z52.32. One additional stand- 
ard on 16-mm print quality control remains to be completed, Ameri- 
can War Standard Method of Determining Printing Loss in 16-Mm 
Sound Motion Picture Prints, Z52.40. A preliminary draft of this 
standard, prepared by Dr. Otto Sandvik and J. A. Maurer, is now 
receiving its final polishing before circulation to the members of the 
Subcommittee B on 16-Mm Sound. It is expected that this proposed 
standard will be one of the most comprehensive documents which has 
dealt with this subject, on which little has appeared in the technical 
literature in the past. 

Another standard in the 16-mm field which is of extreme impor- 
tance to all concerned, is the new American War Standard Specifica- 
tion for 16-Mm Motion Picture Projection Reels and Containers, 
Z52.33-1945. This new standard lists for the first time a comprehen- 
sive set of dimensions and performance requirements for reels which 
will assure a reasonable long service life and provide for proper opera- 
tion on existing projectors. In this connection, I wish to compliment 
D. F. Lyman, Chairman of the Nontheatrical Committee of the So- 
ciety, who served as the Chairman of the Subgroup on Reels, for the 
excellent analysis he made of the dimensional requirements for reels 
and reel spindles necessary for the proper projection of 16-mm sound 
film. Mr. Lyman's detailed study formed the firm engineering 
basis for the work of the War Committee. Incidently, a new Ameri- 
can War Standard for Reel Spindles for 16-Mm Motion Picture Pro- 
jectors, Z52. 34-1945, has also resulted from the work of the Subgroup. 

In the 35-mm field, one new standard has been approved since my 
last report. That is the all-important War Standard for Sound 
Records and Scanning Area of 35-Mm Sound Motion Picture Prints, 
Z52.36-1945. This standard was recommended to the War Com- 
mittee following much work by both the technical committees of the 


Society and of a committee of the Research Council of the Academy 
of Motion Picture Arts and Sciences. As you know, in the past there 
was no complete agreement between all branches of industry on the 
proper dimensional standards for 35-mm sound film. It is hoped that 
this War Standard, when considered in connection with certain war- 
time developments in the art of making raw stock, may result in a 
firm agreement on standards for post-war use. 

The subject of proper tests for judging noise generated by motion 
picture cameras has been one on which there was not too much in- 
formation previous to the war. Subcommittee /, jointly organized by 
the Research Council and the Society, has now drawn up a specifica- 
tion which has been approved as American War Standard Camera 
Noise Test, Z52.60, which should do much to alleviate problems which 
have arisen in the past from the use of noisy cameras. The develop- 
ment of this standard was carried out here in Hollywood under the 
direction of Gordon E. Sawyer. 

Another group of war standards which have been approved in the 
motion picture field are Screens Sizes, Z52.41; Screens Sound 
Transmission, Z52.44; Screens Brightness, Z52.46; and Screens 
Whiteness, Z52.45. These standards contain the essential screen 
requirements for proper projection of motion pictures under both 
Armed Forces and professional conditions. These standards also 
form the basis for another group of 4 specifications for spring-roller, 
springless-roller, folding, and frame-mounted screens which are in- 
tended for eventual adoption as joint Army-Navy procurement speci- 
fications for the Armed Forces. 

Proposed standards for both 16-mm and 35-mm camera viewfinder 
apertures, lens mounting dimensions, and lens registration distances 
are still under development and it is hoped that this work will be 
completed in the next 60 days. 

With the growing amount of direct 16-mm equipment, efforts are 
also being made to crystallize war standards which will stabilize the 
interchangeability situation as far as 16-mm camera film magazines 
are concerned. Draft standards have now been drawn up for the 
400-ft gear-driven magazine and for both 200-ft and 400-ft belt- 
driven magazines. 

A proposed standard nomenclature for optical filters, which is based 
on the spectrophotometric characteristics of the filters, is now nearing 
its final draft stage. As you know, this subject has been one which 
has caused much difficulty in the past since many users of filters are 

36 J. W. McNAIR Vol 45, No. 1 

not quite clear as to the way in which filters affect the picture re- 
corded on the film. In some cases, users have judged filters by their 
spectral appearance rather than by their radiometric characteristics 
with the unfortunate results which might be expected. There has 
been no information available to the industry as to the filters of dif- 
ferent manufacturers which are equivalent photographically from the 
standpoint of their spectrophotometric curves. With the adoption 
of a standard nomenclature, such difficulties will disappear. 

Another project now underway which is of great interest to the 
members of the Society, is for standard synchronization marks for 
both 35- and 16-mm release print negatives and other preprint mate- 
rial. As some of you may know, there have been no accepted stand- 
ards on this subject since the birth of sound on film. Each laboratory 
and producer has used his own peculiar system of identifying editorial 
and projection synchronism for both 35 and 16 mm. With the war- 
time interchange of negatives between producers and the Armed 
Forces, by the Armed Forces themselves, among producers, and 
among processing laboratories, the differing systems of synchroniza- 
tion marks have caused much confusion and loss of time in printing 
and inspection as well as wastage of much film. As the result of the 
circulation of draft standards to the Armed Forces and the interested 
committees of the Research Council and the Society, a number of 
changes have been suggested in the tentative proposal circulated in 
March, 1945. This has now been modified and the new draft is now 
before both the Society and the Council for their recommendations 
on its proposed submission to letter ballot of the War Committee. 

There are several other standards pertaining to motion pictures 
such as those for camera motors, 16-mm warble test films, etc., which 
are still under development. 

I should also like to say a few words as far as the subject of war 
standards for still pictures is concerned, since many members of the 
Society are also interested in these standards. Most important of 
the standards for still pictures to be completed since the last con- 
ference of this Society is the American War Standard Specification 
for Photographic Flash Lamps, Z52.43. This purchase specificatidh 
which has since been adopted as Federal Specification, W-L-122, for 
use by all procuring agencies of the U. S. Government, is the first 
ever to be written on the subject of flash lamps. The method of ap- 
proach used in writing this specification was to rate the lamps on the 
basis of their picture-taking qualities, rather than on the total amount 


of light output, or the peak light output, which so many lamp users 
have attempted to use as a guide in rating lamps in the past. 

Other still photographic standards adopted included one for 35- 
mm slide films which are also used extensively for training and in- 
structional purposes and another for testing the resolution of slide- 
film projector lenses. Proposed standards for still contact printers 
and for slide-film projectors are in the final stages of approval. In 
addition much work has been done on standard test methods for both 
between-the-lens and focal-plane shutters and on the standardization 
of shutter markings. 

As you may judge from the above brief report, the amount of detail 
involved in conducting the work of the War Committee has been quite 
extensive. All in all, 191 drafts alone have been circulated to the 
War Committee and its subcommittees on the 71 projects which have 
either been approved or reached a satisfactory stage to merit assign- 
ment of an identification number. 

The part the Society and its officers played in the work of the War 
Committee has been a large one and without their cooperation much 
of this work could never have been accomplished. The Research 
Council of the Academy has been extremely helpful, particularly with 
regard to the standards for 35-mm motion pictures, since it has helped 
us to a very great extent by assuming the secretarial burden for three 
of the subcommittees which have been active. 

In conclusion, I should like to say that copies of the various Ameri- 
can War Standards, which have been approved, may be obtained 
from the American Standards Association's offices in New York by 
those who have not already done so. Copies of the various draft 
standards, which are under consideration, will be gladly furnished, 
without charge, to anyone who is interested. 



Summary. A description is presented of a new high-pressure mercury arc lamp 
with associated control circuits in which optical inverse feedback is employed to pro- 
duce an adjustable lamp capable of high stability and uniformity of output suitable 
for use in printers. Curves illustrating performance are included and discussed with 
possibilities for use of this lamp in variable-density recording. 


Prior to the development of the hot cathode type of high-pressure 
mercury arc lamps no methods of intensity control or modulation 
were successful because of two inherent characteristics in the mercury 
pool type. The first of these was nonlinear relations between light 
output and power input, and the second, instability caused by the 
movements of the cathode spot on the mercury pool. The cathode 
spot in such arcs is in constant motion because the evaporation of 
mercury under the spot moves it rapidly from place to place and re- 
sults in flutters in the current through the lamp as well as changes in 
the position of the arc stream within the envelope. The light from 
the arc is continually varying. 

With the development of the solid activated electrodes for the high- 
pressure mercury arc lamps the unstable cathode spot was eliminated, 
but the arc still retained the nonlinearity between the input and the 
light output. The light output was not proportional to either the 
current or the power input. 

Fig. 1 illustrates some curves plotted from data taken on a 100-w 
high-pressure capillary type of mercury vapor arc lamp both for the 
starting condition before the lamp was warmed appreciably and for 
the hot condition when it was operating normally. The nonlinearity 
between the light output and the arc current is very apparent and 
indicates that there can be considerable current flow when the light 

* Presented Oct. 16, 1944, at the Technical Conference in New York. 
** Hanovia Chemical and Manufacturing Co., Newark, N. J. 




has practically disappeared. Should an attempt be made to modu- 
late such a lamp it is obvious there would be noticeable distortion 
and the result would be disappointing. 

Another and unexpected defect in the light output from the mer- 
cury arc when applied to recording was the presence of a very high- 
frequency audible hiss which interfered seriously with high-fidelity 
results. The hiss is believed to originate in the mercury molecule it- 
self during the process of ionization. 








' / 




















FIG. 1. Performance curves of a high-pressure mercury vapor 
arc lamp illustrating variations of relative light output for current 
input taken on a cold lamp immediately after starting, and on a 
normally operating one. 

About the year 1926 James R. Balsley conceived the idea of apply- 
ing inverse feedback circuits to the control of the light output char- 
acteristics of mercury arcs. His ,work resulted in U. S. Patent No. 
2,242,638 from which a summary is taken as follows. 

The object was to produce a control circuit which would operate in 
conjunction with a high-pressure mercury arc to produce a lamp 
whose light output remained at a steady level during variations in 
operating conditions that normally resulted in changes in light out- 
put. Such conditions were changes in ambient temperature, drafts 
of air against the arc tube, deterioration of the arc tube, variations 
in the electric power supply, variations inside the arc tube caused by 
varying mercury vapor pressures, etc. 



Vol 45, No. 1 

Mr. Balsley operated a d-c mercury vapor arc lamp having solid 
activated electrodes from a d-c source using triode radio tubes for the 
main stabilizing resistance. All or most of the current for the arc 
lamp passed through these triode tubes and was controlled by them. 
He then placed a photocell at a place where light from the arc lamp 
could fall upon it and by means of an amplifier, the output of the 
photocell was fed back into the control grids of the main triodes. 
The photocell was so connected that an increase in the light reaching 
it caused a response which reduced the input to the arc lamp, while 
a reduction in the light reaching the photocell caused an increase in 
the current reaching the arc lamp. The action of the phototube 


PAT. 2,242,638. 

FIG. 2. Control circuit published by J. R. Balsley in 
U. S. Patent No. 2,242,638. 

therefore was always to reduce or entirely prevent changes in the 
strength of the light reaching it. It could thus hold the light steady 
at whatever level the circuit adjustments permitted. 

The circuit published by Mr. Balsley is shown as Fig. 2, and a re- 
arrangement of it is shown in Fig. 3 to make its operation more 
understandable. There is one feature of this circuit of special note be- 
cause of the variation in potential to ground of the cathode of the main 
control triode. Since the voltage passes through this tube to the arc 
lamp the cathode potential varies with the arc voltage. The grid of 
the main triode is connected in a circuit which does not vary with the 
arc lamp voltage and as a result we experienced some difficulty in main- 
taining stable circuit conditions when experimenting with this circuit. 

It was essential that the performance of the control circuit be 

July, 1945 



known to determine how well it served to perform the main purpose 
of maintaining a steady output of light. We investigated it in the 
following manner. 

The light output of the arc lamp was measured while we varied the 
length of the optical path between the arc lamp and the control photo- 
cell. Since the circuit responded in such a manner as to try to main- 
tain constant light reaching the control photocell when this cell was 
moved toward or away from the arc lamp, the light output of the lamp 
became the variable with distance instead of a varying photocell 
response which is the usual effect. It therefore results that the light 
output of the arc lamp must vary with changing distances between it 



FIG. 3. Redrawn circuit of Fig. 2 to simplify the arrange- 
ment of the parts. 

and the photocell in accordance with the well-known laws of optics 
covering the strength of light at varying distances from a source. If 
the light at the control photocell were to be held constant as the cell 
was moved, the light source would, assuming the inverse square law 
to be operative, have to give out 4 times as much light when the cell 
was moved twice as far away. 

Since the distance from the lamp to the cell was easily measurable, 
a setup was made (Fig. 4) in which the lamp was mounted at a fixed 
location with an auxiliary independent photocell beside it to measure 
its light output. Then the control cell was mounted on a moveable 
car. Measurements were made of the response of the control cell to 
show the variations in light reaching it and also the variations of light 
at the lamp. For a theoretically perfect system, the light at the 



Vol 45, No. 1 

lamp should vary as the square of the distance to the photocell, and 
the light at the control cell remains constant. Fig. 5 illustrates the 
results of these measurements. They show that the light at the lamp 
did not increase enough with the distance changes to the photocell 
and that the light at the control cell fell off markedly with a change 
in distance instead of remaining constant. In other words, the con- 
trol ability of the photocell was unable to maintain the light output 
constant since rather large changes in the light reaching it were neces- 
sary to initiate the control action. The curves given are in no sense 
presented as a critical or exhaustive study since they are only plotted 










FIG. 4. Arrangement of equipment used to determine 
the characteristics of the control circuits. Light variations 
at the lamp were measured with the fixed cell. Light varia- 
tions were determined at the control cell as it was moved 
to various distances from the lamp. 

data taken on our setup of Mr. Balsley's circuits. On another ar- 
rangement with different components the results might have been 

This amount of control was held to be insufficient and new circuits 
were developed to increase the sensitivity of the feedback control. 
Fig. 6 illustrates the new circuits. Two sources of voltage are used 
for stability reasons and both are obtained from rectifiers. The arc 
lamp is connected next to the positive voltage end of the supply with 
the control tubes next to ground in such a manner that the cathodes 
of these tubes were at ground potential. This arrangement per- 
mitted close regulation of the grid potentials at all times. The power 
tube grids were operated by a pair of 6L6 beam power tubes. The 

July, 1945 



control grids of the 6L6's were operated by a single 6SF5 high mu tri- 

Performance curves for this circuit were taken with the same ex- 
perimental procedure used for the other circuits and the results are 
shown in Fig. 7. The light output of the lamp is shown to be very 
near the theoretically correct values for all positions of the control 
cell and the light reaching the control cell varied but slightly with 
distance. Such performance indicated ability of the cell to exercise 
high stabilizing effects through- 
out a wide range of light out- 
put levels. 

Tests of performance were 
also made by the use of an 
oscilloscope. Light from the 
lamp was received upon a 
photocell and connected to 
the oscilloscope to test visu- 
ally the effectiveness of the 
control. Ripples and varia- 
tions that were visible in the 
light when the feedback cell 
was inoperative were seen to 
be greatly reduced as soon as 
the feedback was re-estab- 
lished. Disturbances in the 
light that were introduced 
from external sources were 
compensated for to a maxi- 
mum extent. A test film was 
obtained having a wide variety of densities and was passed between 
the light and the control cell. The circuit response was such that 
the light passing through the various densities of film was maintained 
almost constant, suggesting a possible use of such a control system 
for automatic control of printing processes. 

Tests were made to show the stability of the light levels when they 
were changed from one output to another over a period of time to 
test the ability of the circuit to maintain the light constant. These 
tests indicated that the light could be adjusted within a complete 
range of about 11 to one, be changed instantaneously from one level 
to any other level, and remain steady at the new level. 





























3 '' 


*> ' 








D 6 




FIG. 5. Performance curves of cir- 
cuits set up by Hanovia based on those 
published by J. R. Balsley in U. S. 
Patent No. 2,242,638. 


L. F. BIRD Vol 45, No. 1 


During the development of a satisfactory control circuit it was 
found necessary to develop a new type of mercury arc lamp that 
would be able to operate with the control circuit. The simple form 
of high-pressure mercury arc lamp was not satisfactory for the pur- 
pose. The usual mercury arc lamp is designed to contain a limited 


'SO Vol+s 

FIG. 6. New circuits developed for the control of a 
mercury arc lamp by the use of optical inverse feedback. 

quantity of mercury which is all vaporized when the arc is operating 
normally. The energy dissipated in the arc is depended upon to keep 
the envelope hot enough to maintain the enclosed mercury vaporized. 
If there should be insufficient energy dissipated for this purpose, part 
of the mercury condenses, the voltage across the arc is reduced, and 
the light output goes down. With the arc operating at reduced volt- 
age an excess of current is required to bring" the light output up to 
maximum. The ability of the power supply to furnish current may 
be limited as is the ability of the arc electrodes to carry current, so it 

July, 1945 



is essential that the vapor pressure and the arc voltage be maintained 
at or near the full values. 

For the particular arc lamp used with the control, the arc current 
is rated at 0.4 amp and the voltage drop at about 270 v. When the 
arc currents are above about 0.3 amp, there is sufficient energy to 
keep the voltage up to normal, but as soon as the current drops below 
this value the energy is insufficient and the voltage and pressure fall. 
Since the controlled arc has to be capable of operation at currents as 
low as 50 ma indefinitely, it becomes essential that energy be supplied 
to the arc from some additional 
source. Consequently heater 
coils were added to the struc- 
ture supporting the arc tube in 
such a manner that the vapor 
pressure could be maintained at 
its full value without regard to 
the input to the arc itself. 
These heaters consume about 
50 w and can be supplied from 
any source of power capable of 
delivering about 50 v and one 
ampere. The final lamp is 
shown in Fig. 8. 


'*JV 1 



" r 

















* / 




; / 

z ; 








Modulation of the light from FIG. 7. Performance curves of the 

the arc is accomplished by in- new , circuits for control of a mercury 

arc lamp. 

troducing the modulating volt- 
age into the circuit so that it will have the same effect upon the 
light as if it originated in the photocell. The photocell is then free 
to act as usual and can require the light output to follow quite 
closely the pattern of the modulation voltage. 

Direct electrical modulation of light sources has not received much 
application because of the lack of a suitable light source. The gas- 
type lamps originally employed suffered from short life and from 
poor frequency response at the higher ranges. Modern practice 
employs steady light sources and mechanical light valves for sound 

Our controlled arc overcomes most of the weaknesses that caused 



Vol 45, No. 1 

rejection of the previous lights. It has a long life with slow de- 
terioration and it modulates well throughout the entire audio spec- 
trum with complete stability 
and is capable of high percent- 
ages of modulation with little 

Data were taken to illustrate 
the performance of the arc 
when controlled by a modulat- 
ing voltage. The first curve in 
Fig. 9 shows the variation of 
output light with the feedback 
circuit inoperative. About 75 
v were necessary for complete 
control and there was con- 
siderable curvature from maxi- 
mum to minimum showing that 
there would be considerable dis- 
tortion under these conditions. 
The second curve shows the 
same data taken with the feed- 
back in normal operation. It 
shows that more than 300 v 
are now necessary to produce 
control and that the light out- 
put varies almost directly with 
the applied voltage. Such a re- 
sult is exactly what would be ex- 
pected since it illustrates very 
well the way in which the inverse 
feedback limits the disturbance 
introduced into the circuit, and 
that with it more than 5 times 
as much voltage is required to 
effect the same range of control 
that was obtained without it. 

FIG. 8. 

The Hanovia modulating arc 


In conclusion, it has been shown that the automatic control with 
the special high-pressure arc lamp provides the following : 

July, 1945 



(1) A high-intensity concentrated light source, rich in the highly actinic rays, 
suitable for mounting in a printer. 

(2} An adjustable light that can be pre-set to any level of intensity and which 
will stay steady at that level. 

(5) A light that can be altered from one level of light output to another instan- 
taneously and without an alteration in the quality of the light. 



UT 7 





x^ f 

















00 - 

150 - 

100 1 - 




100 + 


FIG. 9. Performance curves of the Hanovia modulating 
arc lamp and controls showing the effects of the inverse feed- 
back circuits upon the performance and the control of the 
light with varying applied control voltages. 

(4) A light that is adjustable over a range of output of 11 to one to meet the 
need for this range in printers. 

(5) A light that can be operated from the usual 60-cycle supply lines without 
the need for specially regulated power supplies. 

(6) A light that can be modulated with high fidelity to the control voltage 
throughout the audio-frequency range and which has stability with long operat- 
ing life and slow deterioration. 



Summary. The purpose of this article is to trace the history of microphones, as 
used in motion picture production sound recording, and the art of their use. It 
covers also the development of accessory equipment, such as wind guards, rain guards, 
microphone booms, etc. 

The history of microphones in sound pictures may be divided into 3 
periods. According to the name of the moving element in the unit 
used at the time, one may thus speak of the condenser, the moving- 
coil, and the ribbon microphone period. The third period, which is 
the present, is not too clearly defined, however, by the term "ribbon," 
since the present microphone is a complex unit consisting of 2 "rib- 
bons" or a "ribbon" and a moving-coil unit. 

There is, of course, also a future. Since the developments of the 
future are always based on the accomplishments of the past, it may 
serve us to look ahead, and after having enumerated our wants, 
take inventory to learn how far we have progressed. And since we 
believe in thoroughness, we went out and gathered all the needs and 
desires and hopes of practically everyone connected with the record- 
ing of sound. This is what we found : 

(1} The microphone- to-come should be practically invisible. 

(2) It should be capable of being moved by invisible means. 

(5) If it must be made of visible materials, it should be no larger and weigh no 
more than a plum, at most no more than a lemon. 

(4} It should introduce so little distortion that our present measuring equip- 
ment cannot measure it. 

(5) It should be automatically directional, that is, pick up in each scene only 
the desired sounds and ignore all others. 

(6) It should not be affected by wind, rain, and snow. 

(7) It should record the sounds produced by the weather conditions only when 
the mixer so desires. 

* Warner Bros. First National Studios, Burbank, Calif. 
** RCA Victor Division, Radio Corporation of America, Hollywood. 



The above is truly optimistic and may sound facetious, but it rep- 
resents the perfect microphone. We doubt, however, whether there 
is a further cry between the microphone-to-be and the present units 
than there is between the earliest microphone and what we employ 
now. Only those who labored in the early days of sound recording 
can properly appreciate the advancements in this art, and can chuckle 
when thinking of the bygone days. 

Let us therefore visit an early "set," say sometime in the summer of 
1929, after the Jazz Singer had begun to outline the shape of things 
to come. There was our matinee idol dressed to kill and with fire 
in his eyes. He enters a room, stops, speaks a few lines .with head 
raised slightly, and then walks to a spot designated by chalk marks 
on the floor, there to deliver the remainder of his lines. The proce- 
dure seldom varied. The heroine's deportment suffered similarly. 
The director suffered, too, dissimilarly. 

Why all these stilts, the new-timer may well ask, still conscious 
perhaps of the finished job his favorite movie star executed the night 
before. The answer is short, and may be termed "microphonics.'' 
It was at the time in no way related to the microphone used on the 
set. Whatever originated in the microphone and the sounds from it 
were hideous microphonics did not come from there. The generator 
of microphonics was a vacuum tube a temperamental wire mesh 
inside an evacuated glass bulb. It spat and sputtered at the slightest 
touch or motion and sometimes apparently for no reason at all. 

However, then as now, time marched on. If it was fraught with 
danger to move the amplifier with its supersensitive vacuum tube, 
some one said, why not move the condenser microphone attached to 
the amplifier? The actor would thereby at least be able to turn 
around, instead of remaining glued to the floor, at the spot marked 
X . Any action was better than none. 

Thus, the actor was able to walk about slightly. The increased 
radius of action was from only 3 to 5 ft the length of the flexible 
gooseneck which ran from the amplifier to the microphone. In this 
way the periodically paralyzed actor began to amble once more, and 
the director heaved a (small) sigh of relief. 

Since this article intends to trace the history of microphones and 
the art of their use, it may be well not to divorce this unit from its 
suspension, but to discuss the two side by side, as it were. In this 
early age of the condenser microphone we have seen that the amplifier 
was hung in some fashion from the ceiling and the microphone was 

50 W. A. MUELLER AND M. RETTINGER Vol 45, No. 1 

connected to this electrical equipment by means of a flexible goose ; 
neck. Unfortunately it was standard motion picture practice at that 
time to photograph the long-shot, the medium-shot, and the close-up 
simultaneously from several different cameras each housed in a 
separate booth to prevent the camera noise from disturbing the re- 
corded sound. This procedure frequently showed the microphone 
in the long-shot, and it became almost an art in itself to hide the unit 
unobtrusively among the props of the set. In fact, one of the essen- 
tial talents of a good mixer was his ability to place the microphone 
so that it would not be visible in the picture but would yet record the 
dialogue naturally and intelligibly. This, of course, was a serious 
problem. Concealment of the unit in a vase, inside clocks, ash-trays, 
or underneath lamp shades frequently resulted in no little damage 
to the recorded sound because of sound shadows, cavity resonances, 
and other undesirable conditions created by these stealthy micro- 
phone locations. Still, many lines so recorded were surprisingly 
good not equal to the best of today perhaps, yet clear enough so as 
not to detract the listener's attention or to tax his ears unduly. 

It may be well at this point to look at the condenser microphone 
itself a little more closely. Its chief impediment was the noise pro- 
duced in it by moisture condensation. This condition could bring 
forth in it the most exasperating crackle of fire as if a miniature hell 
had broken loose in it, with the result that every one was tormented 
accordingly. To relieve this condition the microphone, when not 
in use, was usually kept in a desiccator filled with calcium chloride 
to absorb the moisture from the little spit-fire. 

Again time marched on. By and by condenser microphones were 
constructed which were quieter and more trustworthy in their be- 
havior. Also, the microphonic vacuum tube became less micro- 
phonic, and before long a heavy type of boom, practicable if primi- 
tive, made its appearance on the stage floor of many of the studios. 
Thus disappeared the many wires and ropes which had been used to 
suspend the microphone or amplifier, or both, and had reminded one 
of the strings in a marionette theater, and with them went many a 
sigh of relief, heaved by the mixer as well as the "grips" and "juicers," 
not to forget the cameraman. 

We have spoken several times of a mixer, taking it for granted that 
everyone knows what the word means. Technically, the word really 
has two meanings. So far we have used it only to denote the man who 
listens to spoken lines over a set of headphones to gain a peremptory 


impression of the sound recorded on the film. The second meaning 
of the word is that of the "mixing console," or the small table with 
its several volume controls and electric meters. To be explicit, how- 
ever, we shall speak of the mixer only as the man, and write out mix- 
ing console when referring to the portable collection of dials and 

To the mixer, in the condenser microphone period, were ascribed 
special functions, some illusory and some real, and he was housed 
accordingly usually in a small booth carrying a double pane window. 
It may be of interest here to quote a paragraph in an early paper on 
sound recording, written for the purpose of acquainting the personnel 
of the new art with the most satisfactory manner of recording tech- 

"Let us watch the monitor at work. Posted behind the glass 
windows, he sees the artists, and when they speak, hears them in- 
directly through the medium of the control loudspeaker. He has, 
therefore, thetllusion of being near them, although actually separated 
by many thicknesses of glass which insure silence. The scene is re- 
peated. He listens, and communicates his impressions by micro- 
phone and loudspeaker to the director. He points out any crackling 
noises or unpleasant intonations, increases or decreases the sound in- 
tensity, modifies the orientation of the microphones, and indicates 
any acoustic anomalies produced. His control, which is exercised 
through the medium of a microphone, is much more critical than that 
of the director in the studio." 

This was of yore. Today the mixer sits with the director behind 
the cameras, listening to the recorded dialogue by means of ear 
phones while his expert fingers move the volume dials on his mixing 
console. The microphone and microphone boom have been improved 
to such an extent that the director is no longer hampered in any way 
as regards the movement of the actors. He stages the scenes natu- 
rally, knowing that the mixer will be able to get good sound pickup 
regardless of how complicated the shot is. The mixer has become 
more of a consultant to the director on the mood, tempo, and loud- 
ness of the lines spoken by the actor. 

Comes now the era of the moving-coil and moving-conductor mi- 
crophone. These units could be divorced from their associated ampli- 
fiers by 50 ft or more of cable and increased the radius of action 
manifold. Now the actors could wander about at will again and the 
chalk marks were erased from the floor. The microphones, moreover, 

52 W. A. MUELLER AND M. RETTINGER Vol 45, No. l 

were attractively curvilinear, as if to validate the modern artist's 
slogan that "form follows function." 

As the art of microphone construction advanced, so did the art of 
microphone boom manufacture. Booms were built which were small 
marvels of mechanical ingenuity, with telescoping duralumin arms 
and noiselessly controlled swivel arrangements. They rested on 
rubber-tired wheels, and could be swung about and raised and lowered 
and made nary a sound. Another milestone has been removed from 
the road which was apparently being traversed with seven-league 

But as most roads, so this one, too, it was found, had its turns and 
encumbrances. The early sound pictures were all made in special 
soundproof stages until one producer presented "the first outdoor 
talking picture." Then sound pictures moved to the desert, the 
mountains, and the seashore, and many new problems were presented 
for solution to the sound engineers. One of the worst of these was 
wind noise which sounded through the microphone lil distant artil- 
lery fire instead of a gentle zephyr. 

Early attempts to reduce the undesirable wind noises on exterior 
sets consisted essentially of wrapping one or more layers of cloth 
around the microphone. This, of course, made for greatly distorted 
and very unnatural sound which was barely intelligible. Still, it gave 
the picture some semblance of realism, and was better than no sound 
at all. Soon wind screens were made as an attachable device crude 
in form and primitive in execution, but superior to the miniature 
gunny sacks or handkerchiefs used before. As covering for the metal 
skeleton a variety of cloths were employed voiles, silk, cheesecloth, 
and even burlap usually applied to the outside of the screen. 

When rain and snow scenes were to be made, a protective device 
known as rain screen was attached to the microphone. It consisted 
chiefly of, first, a fine mesh metal screen to break the drop into smaller 
particles, second, a layer of felt below the metal mesh onto which 
these small particles could drop without a sound, and third, under- 
neath the felt a layer of oilcloth or other material through which 
water could not penetrate. This device is still used extensively in 
the industry. 

The wind screens of today, however, are finished pieces of alumi- 
num handicraft, easily attachable, and well rounded to avoid any 
whistling effect. They are still large, however, and not infrequently 
will throw an annoying shadow. But who can judge their future size 


without judging the size of the future microphone at the same time? 
There is a relationship between the two, and as the microphone be- 
comes smaller, the wind screen may contract likewise. 

Comes now tjie third period in our microphone history that of the 
ribbon unit. For a long time it has been felt that if a microphone 
could be constructed which would record only sound coming from a 
certain direction and ignore sound arriving from other directions, 
many unwanted sounds, such as those from motor-driven cameras, 
arc lamps, camera "dollies," and associated equipment in motion on 
the set during a scene, could be suppressed. 

There was, of course, nothing new in the idea of constructing a di- 
rectional sound pickup device, as testified by the fact that so-called 
parabolic sound concentrators had been used in the field fairly early. 
Their drawbacks lay in their large size, heavy construction, difficulty 
of operation, and the fact that they exhibited numerous and sharp 
resonances which marred the quality of the recorded sound in no 
minor manner. For special operations, however, where intelligibility 
or naturalness of speech could to a degree be sacrificed, they proved 
very valuable, and may occasionally still be seen. Indeed, as far 
as the construction of a device is concerned which can exhibit an ex- 
tremely narrow angle of pickup, the parabolic sound concentrator 
has no rival, as shown by its former use as an aircraft detector. 

Credit for building a unidirectional microphone which could be 
suspended from a boom goes to three men Olson, Weinberger, and 
Massa who described such an instrument in the Journal of the A cous- 
tical Society in October, 1933. But here as elsewhere it was a far step 
between the first model and the present unit. Suffice it to say that 
the early microphones were so large and heavy that few of them were 
used regularly on production. Indeed, it has been only during the 
last 3 years that this type of microphone, variously known as the 
unidirectional or cardioid, has been employed extensively by the in- 
dustry. It may shrink further and become still lighter and who can 
tell but that in a few years it may be the size of a lemon, and but a few 
years further hence, the size of a plum? The engineer has devious 
ways his plans to achieve. 



Summary. As the result of Army complaints on the printing quality of Van- 
dykes, or brown line print master negatives, aircraft companies have spent consider- 
able time and effort in expensive tracing heavy-up work. 

Lockheed has specified and had prepared a new and different type of Vandyke paper 
having sufficient latitude and contrast to produce acceptable prints from combination 
pencil and ink line tracings. The paper carries a slow photographic emulsion, is 
processed on the standard blueprint machines under normal operating conditions, and 
can be furnished at the approximate cost of the present Vandyke paper. 

In order to speed production, it is customary for war plants to pre- 
pare direct pencil drawings of parts or structures on a thin tracing 
paper suitable for use in blueprint reproduction. With the extended 
distribution of war materials, it became necessary for the Armed 
Forces to prepare copies of certain of these prints. Specifications 
called for the preparation of a master negative as a Vandyke, suitable 
for use in the contact printing of blue line positives. 

The method of preparing Vandykes, or master negatives of tracings 
for subsequent blue line reproduction, involves the following: 

(1) The original drawing or tracing is prepared with pencil on tracing paper. 
If the lines are sufficiently heavy, a Vandyke can be made by direct contact 
printing, however, a copy cloth photographic-type positive is often made from 
this tracing to gain contrast, minimize spots and dirt, and provide a durable rec- 
ord for use in the Engineering Department. This print is often corrected with 
either ink or pencil lines and sometimes both. 

(2) A Vandyke negative is made from the copy cloth positive. The Vandyke 
consists of a pretransparentized 100 per cent rag, 12-1 6-lb paper base coated 
with silver salts and suitable halide acceptors such that following exposure to 
light and treatment in a sodium thiosulfate solution, the exposed sections are 
converted to brown silver or silver sulfide. 

A print made in the above manner from a copy cloth bearing ink 
and pencil corrections fails to reproduce all lines sufficiently different 

* Presented Oct. 17, 1944, at the Technical Conference in New York. 
** Lockheed Aircraft Corporation, Burbank, Calif. 



from the background so that acceptable blue line prints can be made. 
The problem is well illustrated by the first 2 figures circulated by the 
Engineering Department. 


FIG. 1. 

The first efforts of Research were directed toward the development 
of an improved type of drafting pencil to more closely duplicate the 
quality of ink lines. The immediate results of this work were nega- 
tive. Attention was then directed toward utilization of chemical and 
photographic principles in the specification of a Vandyke paper of 
such quality that the backlog of combination ink, silver, and pencil 



Vol 45, No. 1 

line tracings could be adequately reproduced without requiring heavy - 
up work. 

To review briefly the principles governing reproduction in light- 
sensitive materials, it will be recalled that two men, Hurter and 

FIG. 2. 

Driffield, suggested the following basis for photographic sensitometry : 

The opacity of a material can be expressed in terms of the incident 

light intensity required on one side in order to transmit light of unit 

intensity = /o//. The transparency of a material is a measure of the 


light getting through, T = Density, first studied in terms of silver 

concentration per unit volume, was established as the logarithm to the 

base ten of the opacity = logio - = logio 

The properties of the photographic emulsion as the result of con- 
trolled incident energy or exposure, depend upon the treatment of 
the silver salts during preparation, the gelatin protective colloid, 
the type of development, and many other factors. Fig. 3 illustrates a 
typical characteristic curve of a photographic emulsion showing the 
relation between density as defined above and standardized exposure 
to white light as obtained on an Eastman sensitometer, type lib. 
As this exposure is increased to give a perceptible density for given 
developing conditions, the density varies almost linearly with expo- 
sure in the section known as the toe of the curve. A print in this 
density region would be underexposed and thin with little difference 
between densities. Above this toe portion is the straight line portion 
wherein density varies linearly with log exposure. This is a most 
useful section of the characteristic curve because the opacities of the 
silver image are proportional to the exposures produced by the cor- 
responding brightness of the subject. The shoulder section of the 
curve represents a region of almost constant density with increased 
exposure. A print in this section would be overexposed and dense 
with little difference between densities. A study of Fig. 3 will show 
that the straight line portion of the curve can be represented by the 
equation : 

Density = y (logw Exposure + K} 

The factor 7 is a measure of the slope of the straight line portion 
and is thus a measure of contrast. Contrast is a function of manu- 
factured stock characteristics and color temperature of exposing light, 
but if the image be placed in the region of correct exposure, contrast 
can be altered by development. The length of the straight line por- 
tion is termed the latitude of the stock and again is a function of 
manufacture, principally emulsion thickness. The ratio of densities 
in the region of correct exposure, then, is determined by subject con- 
trast; the differences between densities are determined by develop- 
ment. It is common in the photographic industry to utilize these 
principles of sensitometry in controlling the speeds, latitudes, and 
contrasts of photographic materials. 



Vol 45, No. 1 

July, 1945 








Vol 45, No. 1 

From these fundamentals it is immediately evident that if a stock 
had sufficient latitude it would be possible to print the heavy ink 
lines, the light pencil lines, and the thin base to produce definite 
density ratios. If, through contrast, sufficient density difference be- 
tween pencil line and background be available, then acceptable blue 
line prints could be made. 

The vendor of the present Vandyke paper has been unable to supply 
either greater latitude or higher contrast paper. Chemical intensifi- 

FIG. 5. Processed sensitometric exposures. 

cation of this Vandyke image after processing, while possible, was 
considered impractical. 

In order to determine the density range for reproduction, the dif- 
fuse transmission densities of the lines of a typical copy cloth were 
determined on a Capstaff densitometer, and are listed substracting 
the cloth or base density of 0.38. Typical values are as follows: 

Cloth Base 


The investigation of photographic, paper in comparison with Van- 
dyke paper disclosed several interesting facts concerning reproduction 
of these density differences. 

July, 1945 



I ' 1 B 





Vol 45, No. 

A sensito metric-type exposure was made on the Vandyke paper 
under processing machine conditions, a similar exposure was made on 

FIG. 7. Graphic reproduction summary. 

a 12-lb base photographic paper prepared by the vendor at the re- 
quest of Research. The diffuse densities of the processed tests were 


determined and plotted against the corresponding log exposure 
values. Fig. 4 shows the characteristic curves for the 2 papers 
shown in Fig. 5. The print density is plotted against the relative 
exposures obtained by printing through the densities of Fig. 3. It 
can be seen that if the base or clear sections of the copy cloth are 
printed to maximum density in each case, then the resulting print 
density differences are : 

Vandyke Photo 

Between background and pencil lines . 34 1 . 02 

Between background and silver lines 1 . 10 2.36 

Between background and ink lines 1 . 10 2.36 

It is evident then that using this photographic paper there is a 
density difference between background and pencil lines of 1.02, or 3 
times the density difference available using Vandyke paper. Like- 
wise, because of greater contrast, the maximum density of the photo- 
graphic paper background compared to a white line is 2.36 compared 
to 1.10 for the Vandyke paper. This, of course, will result not only in 
better Vandykes but also in better blue line prints because slower 
printing speeds can be used without causing exposure through the 
background. The prints in Fig. 6 were made from a tracing rejected 
by the inspector. The two on the right represent the best Vandyke 
negative available from the pencil tracing and the best .blue line print 
that could be made therefrom. The corresponding photographic 
Vandyke and positive blue line print are shown on the left. It can 
easily be seen that an acceptable print is available. 

A graphic summary of this type of reproduction is shown in Fig. 7. 
The upper left quadrant represents the range of subject contrast en- 
countered in a typical tracing. The characteristic curves of the Van- 
dyke and photographic papers are shown in the lower left quadrant. 
Light passing through these densities records on the blueprint paper 
as shown by its characteristic curve in the lower right quadrant. Al- 
though reproduction of the original tracing is sufficient, it is highly 
desirable to present all lines of maximum density on a clear back- 
ground. This condition is illustrated by the dotted line in the upper 
right quadrant. By tracing the subject brightness or effective ex- 
posure of any point through 270 degrees to compare initial and final 
densities, it can be seen that the silver and ink lines record to repro- 
duce maximum blue line density whereas the pencil line is about half 
maximum density. The greatest difference, however, is noticeable 


between the background and pencil line densities; in the case of the 
Vandyke, the density difference is 0.12; on the photographic paper 
it is 0.50. It can readily be seen that the master negative density 
differences which control the extent of blue line printing can be in- 
creased fourfold by the use of photographic paper. 

Although certain slight modifications of the submitted paper are 
being requested, very pleasing results have been obtained with a 
photographic emulsion on a 12-lb base which, although not pre- 
transparentized, is characterized by the same printing speed as brown 
line or Vandyke paper. The material is quite safe to reasonable use 
under incandescent illumination; it will fog within a few seconds 
under fluorescent lights. The image develops within the first 15 sec 
of the 30-sec development in D-72. There is no appreciable gain in 
density or contrast with prolonged development. Hardening hypo 
fixes the paper within 90 sec; this is followed by a 2-min wash and 
normal drying. The cost is comparable to that of the present Van- 
dyke paper. 

West Coast aircraft plants have evidenced considerable interest in 
this new product developed by Research to solve a Lockheed problem. 
A material capable of the reproduction discussed above and supplied 
at a price comparable to current brown line papers will have universal 
application in the blueprint and photo-sensitive reproduction indus- 



The editors present for convenient reference a list of articles dealing with subjects 
cognate to motion picture engineering published in a number of selected journals. 
Photostatic or microfilm copies of articles in magazines that are available may be 
obtained from The Library of Congress, Washington, D. C., or from the New York 
Public Library, New York, N. Y., at prevailing rates. 

American Cinematographer 

26 (Apr., 1945), No. 4 
An All-Friction Drive for Developing Machines (p. 122) 

26 (May, 1945), No. 5 

The Adel Color Camera and Surgiscope (p. 152) 
Television and Motion Pictures (p. 158) 
Post -War Motion Pictures (p. 160) 

26 (June, 1945), No. 6 

A Method of Film Conservation in Motion Picture Pho- 
tography, Processing and Reproduction (p. 188) 
Harman Unveils News Animaction Unit (p. 190) 
Kine-Micrography in Biological Research (p. 192) 
The Television Camera (p. 193) 
.Movement in Movies (p. 194) 

New Repeating Flash Tube for Night Aerial Photography 
(p. 208) 

Better Management (Sec. of The Exhibitor) 

8 (May, 1945), No. 5 

Air Sterilization with Ultraviolet Light Will Give Thea- 
ters True Air Conditioning (p. 5) 
8 (June, 1945, No. 6) 

Transmission Films Added to Projection Lenses Promise 
Clearer, Reflectionless Screen Images (p. 5) 

British Kinematograph Society, Journal 

8(Jan.-Mar., 1945), No. 1 

Future Technical Progress in the Film Industry (p. 2) 
Kine-Micrography in Biological Research (p. 8) 
X-ray Kinematography (p. 9) 

Operational Kinematography in the R. A. F. (p. 10) 
The Work of the National Film Library (p. 13) 


25 (Mar., 1945), No. 3 

A Television Studio Installation Designed for Research 
and Instruction (p. 33) 

W. G. C. Bosco 
W. G. C. Bosco 



W. G. C. Bosco 



A. G. D. WEST 







Electronic Engineering 

17 (Apr., 1945), No. 206 
Scanning Systems for Colour Television (p. 456) 

Electronic Industries 

4 (Apr., 1945), No. 4 

Multi-Channel Sound Recording on Film (p. 92) 
RCA Reveals Projection Television (p. 95) 
DC Picture Transfer (p. 106) 

4 (June, 1945), No. 6 
DuMont's Projection Television (p. 97) 

International Photographer 

17 (Mar., 1945), No. 2 
The Television Cameraman (p. 11) 
Specifications for Photographic Flash Lamps (p. 22) 
Television Topics (p. 24) 

17 (Apr., 1945), No. 3 
The Garutso Optical Balance (p. 13) 
Columbia's Still Laboratory (p. 15) 

17 (May, 1945), No. 4 
Pal's Pals (Puppets) (p. 7) 
Television Topics (p. 23) 

International Projectionist 

20 (Mar., 1945), No. 3 
The Operation and Maintenance of a Popular 16-Mm 

Projector (p. 7) 
Projection Angle Rule (p. 10) 
Proper Operation of Projection Arc Lamps (p. 12) 
Projectionists' Course on Basic Radio and Television 
Pt. 9 (p. 13) 

20 (Apr., 1945), No. 4 

Step-by-Step Analysis of a 16-Mm Amplifier (p. 7) 
Design of a Sound-Level Indicator (p. 10) 

Projectionists' Course on Basic Radio and Television 

Pt. 10 (p. 18) 

20 (May, 1945), No. 5 

Technical Problems of Arc Lamp Design (p. 7) 
Providing Auxiliary Sound Requirements for Motion 

Picture Theaters (p. 9) 
Television Developments (p. 16) 
Projectionists' Course on Basic Radio and Television 

Pt. 11 (p. 18) 


2 (May, 1945), No. 4 
Video Lighting Problems (p. 4) 
Television Station Design (p. 7) 



















Dr. A. F. Turner of the Scientific Bureau of the Bausch and Lomb Optical 
Company, Rochester, N. Y., addressed members and guests of the Atlantic Coast 
Section of the Society at the meeting on May 23. Dr. Turner, long associated 
with the development of nonreflecting films, spoke on "Performance of Coated 
Lenses." The paper covered in a popular manner the optical principles involved 
in the action of the nonreflecting films, the manner in which they are deposited 
commercially i and the results to be expected from them. 

One of the most interesting demonstrations given by Dr. Turner in the course 
of his talk involved the transmission of an image through a series of uncoated glass 
plates as well as through a similar series of coated glass plates. The effect of 
coating was clearly shown by the brighter image obtained and by the absence of 
ghosts, which were clearly visible in the transmission through the uncoated plates. 

The meeting, held in the Roof Garden of the Hotel Pennsylvania, New York, 
was attended by approximately 150. 

High-speed motion picture photography was the concluding subject in the 
Spring 1945 series of Atlantic Coast Section meetings, which was held on June 13 
in the Georgian Room of the Hotel Pennsylvania. The speaker was Henry M. 
Lester, cinephotographer of New York, whose topic was "Continuous Flash 
Lighting: An Improved Source for High-Speed Motion Picture Photography." 

High-speed motion picture cameras, capable of taking pictures on continuously 
moving film at the rate of upward of 3000 frames per sec, produce individual ex- 
posures of the order of 1 /i5,ooo sec and less. Since the actual time during which 
high-intensity illumination is required seldom exceeds one second, the light of cer- 
tain Photoflash lamps will provide satisfactory illumination for this purpose, when 
such lamps are operated in suitably designed equipment. 

Mr. Lester demonstrated his continuous flash lighting unit which provides 
adequate illumination for both black-and-white and color high-speed motion pic- 
ture photography, and showed motion pictures of the equipment in action. 

C. R. Keith, chairman of the Section, announced that the Fall series of meetings 
would cover a wide range of subjects of current interest, the first of which is ten- 
tatively scheduled for September 19. 


The 57th Semi-Annual Technical Conference held at the Hollywood-Roosevelt 
Hotel, Hollywood, California, on May 14-18, inclusive, was one of the most suc- 
cessful meetings ever held by the Society. The quality of the papers presented 
and the wide range of topics discussed were responsible for drawing a large local 
attendance during the 5-day meeting. 



It was decided before the Conference to withhold distribution of the usual ten- 
tative program in order not to promote out-of-town attendance, in cooperation 
with governmental regulations on transportation and hotel accommodations. It 
is felt, however, that many members of the Society would be interested in having 
a list of the papers presented. The Board of Editors therefore is publishing in 
this issue of the JOURNAL the complete program as followed. 

In order to make the information contained in these papers available to mem- 
bers as soon as possible, the Board of Editors is planning to publish this material 
as early as manuscripts on hand permit. 

Monday, May 14, 1945 

12:30 p.m. California Room: SMPE V-E Luncheon, for members and 

guests. D. E. HYNDMAN, President of the Society, presided. 
Address of Welcome: WALTER WANGER, President, Academy of 

Motion Picture Arts and Sciences. 
Guest Soloist: OLGA SAN JUAN, Paramount Singer and Dancer. 

2:00 p.m. Studio Lounge: Afternoon Session. 
L. L. Ryder, Chairman 
J. G. Frayne, Vice- Chairman 

Session opened with a 35-mm motion picture short. 

Report of Convention Vice-President, W. C. Kunzmann. 

"Engineering Committee Reports on Laboratory Practice, Non- 
Theatrical Equipment, Preservation of Film, Television, Stand- 
ards, Theater Engineering," by J. A. Maurer, Engineering Vice- 
President, SMPE. 

"Wave Propagation and Outdoor Field Tests of a Loudspeaker Sys- 
tem," by F. L. Hopper and R. C. Moody, Western Electric Co., 
Inc., Hollywood, Calif. 

"Film Noise Spotter," by J. P. Corcoran, Twentieth Century-Fox 
Film Corp., Beverly Hills, Calif. 

"Reverberation Chambers for Rerecording," by M. Rettinger, 
RCA Victor Division, Radio Corporation of America, and James 
Stewart, RKO Radio Pictures, Hollywood, Calif. 

"The Comparison of Beam Power and Triode Tubes Used in Power 
Amplifiers for Driving Loudspeakers," by J. K. Hilliard, Altec 
Lansing Corp., Hollywood, Calif. 

"A New Carbon for Increased Light in Studio and Theater Projec- 
tion," by M. T. Jones, R. J. Zavesky, W. W. Lozier, National 
Carbon Company, Fostoria, Ohio. 

"Airborne Sound Recorders," by G. C. Brubaker, Memovox, Inc., 
Beverly Hills, Calif. 

A demonstration was given. 


8:00 p.m. Studio Lounge: Evening Session. 

K. F. Morgan, Chairman 
C. R. Keith, Vice- Chairman 

Session opened with a 35-mm motion picture short. 

"Progress Report on War Standards for Photography and Motion 
Pictures," by J. W. McNair, American Standards Association, 
New York. 

"Push-Pull FM Circuit and Its Application to Vibratory Systems," 
by Alexis Badmaieff, RCA Victor Division, Radio Corporation of 
America, Hollywood, Calif. 

"Cinemicrography in Biological Research," by R. McV. Weston, 
Secretary, Association for Scientific Photography, Hound wood, 

"An Integrating Voltmeter for Measurement of Fluctuating Volt- 
ages," by Harold Haynes, RCA Victor Division, Radio Corpora- 
tion of America, Indianapolis, Ind. 

A demonstration film was presented. 

"A New Two-Way Theater Loudspeaker System," by J. B. Lansing 
and J. K. Hilliard, Altec Lansing Corp., Hollywood, Calif. 

"A Public Address and Sound Re-Enforcement System Using the 
Duplex Loudspeaker," by J. B. Lansing, Altec Lansing Corp., 
Hollywood, Calif. 

Tuesday, May 15, 1945 

Open Morning. 

2:00 p.m. Studio Lounge: Afternoon Session. 
W. V. Wolfe, Chairman 
N. L. Simmons, Vice- Chairman 
Session opened with a 35-mm motion picture short. 
"A Multisection Rerecording Equalizer," by W. L. Thayer, Sound 

Dept., Paramount Pictures, Inc., Hollywood, Calif. 
"A Three-Band Variable Equalizer," by L. D. Grignon, Twentieth 

Century-Fox Film Corp., Beverly Hills, Calif. 
"Mathematical Analysis of a Recorder Film Drive Filter System," 

by J. R. Alburger, Electrical Research Products Division, West- 
ern Electric Co., Inc., Hollywood, Calif. 
"Improved Film Recording Channels," by Dr. C. R. Daily, Sound 

Dept., Paramount Pictures, Inc., Hollywood, Calif. 
"Organizing 16-Mm Production," by Lloyd Thompson, The Calvin 

Company, Kansas City, Mo. 
"Variable-Area Release from Variable-Density Original Sound 

Track," by J. P. Livadary and S. J. Twining, Columbia Pictures 

Corp., Hollywood, Calif. 
"Automatic Recording of Photographic Densities," by J. G. Frayne 

and G. R. Crane, Electrical Research Products Division, Western 

Electric Co., Inc., Hollywood, Calif. 


8:00 p.m. Studio Lounge: Evening Session. 

Herbert Griffin, Chairman 
G. E. Sawyer, Vice- Chairman 
Session opened with a 35-mm motion picture short. 
"The Projection Life of Film," by R. H. Talbot, Eastman Kodak 

Co., Kodak Park, Rochester, N. Y. 
"Report of the Subcommittee on Projection Sprocket Design," by 

J. A. Maurer, Engineering Vice-President, SMPE. 
"Television vs. Motion Picture Practices," by Klaus Landsberg, 
Director of Television, Station W6XYZ, Television Productions, 
Inc., Hollywood, Calif. 

"Film The Backbone of Television Programming," by R. B. 
Austrian, Executive Vice-President, RKO Television Corp., New 

Wednesday, May 16, 1945 

Open Morning. 

2:00 p.m. Studio Lounge: Afternoon Session. 
H. W. Moyse, Chairman 
S. P. Solow, Vice- Chairman 

Session opened with a 35-mm motion picture short. 
"Positive Vari-Focal View-Finder for Motion Picture Cameras," 
Dr. F. G. Back, Research and Development Laboratory, New 

The following 5 papers were presented by members of the U. S. 

Army Signal Corps Photographic Center, Long Island City, N. Y.: 

"The Psychological and Technical Considerations Employed in the 

Bucky Sound Reproduction and Public Address System," by 

Lieut. P. A. Bucky. 

"An Optical Cueing Device for Disk Playback," by Major G. C. 


"Army Film Distribution and Exhibition," by Major R. A. Kissack, 

A demonstration film was presented. 

"A System of Lens Stop Calibration by Transmission," by Em- 
manuel Berlant. 

A demonstration was presented. 

"Development of Two Automatic Follow-Focusing Devices for 
Use in Cinematography," by Capt. J. T. Strohm and Capt. W. 
G. Heckler. 

A demonstration film was presented. 

5:30 p.m. California Room: A social hour with the Board of Governors. 
Invitations were issued to registered members and guests. 

Open Evening. 


Thursday, May 17, 1945 

Open Morning. 

2 : 00 p.m. Studio Lounge: Afternoon Session. 
Emery Huse, Chairman 

Lt. Comdr. Franklin Hansen, Vice- Chairman 
Session opened with a 35-mm motion picture short. 
"A New Photographic Developer for Picture Negative," by J. R. 

Alburger, Hollywood, Calif. 
"A Note on Chemical Drag Observed with Variable- Density Sound 

Track," by Dr. D. Meschter, Photo Products Dept., E. I. duPont 

de Nemours & Co., Inc., Parlin, N. J. 
"duPont Fine-Grain Sound Films, Type 232 and 236," by H. W. 

Moyse, Photo Products Dept., E. I. duPont de Nemours & Co., 

Hollywood, Calif. 
"The Work of the Naval Photographic Services Depot," by Lt. 

Comdr. F. M. Hearon, Officer-in-Charge, Naval Photographic 

Services Depot, Hollywood, Calif. 
"Colored Trace Oscillographs," by L. S. Trimble and F. W. Bowden, 

Lockheed Aircraft Corp., Burbank, Calif. 

8:00 p.m. Studio Lounge: Evening Session. 

E. A. Williford, Chairman 
L. E. Clark, Vice- Chair man 

Session opened with a 35-mm motion picture short. 
" Practical Utilization of Monopack Film," by C. G. Clarke, Director 
of Photography, Twentieth Century-Fox Film Corp., Beverly 
Hills, Calif. 

"The Printing of 16-Mm Kodachrome Duplicates," by R. M. 
Evans, Research Laboratories, Eastman Kodak Company, 
Rochester, N. Y. 

"Machine Processing of Ansco Color Film," by J. L. Forrest, Ansco, 
Binghamton, N. Y. 

A demonstration film was presented. 

Friday, May 18, 1945 
Open Morning. 

2:00 p.m. Studio Lounge: Afternoon Session. 
W. A. Mueller, Chairman 
P. E. Brigandi, Vice- Chair man 
Session opened with a 35-mm motion picture short. 
"Orthoscope Lenses," by Hal Huff, H & H Optics, Los Angeles, 

"An Automatic Interlock Switch," by D. J. Bloomberg, Republic 

Pictures Corp., Studio City, Calif. 

"A Discussion of the Acoustical Properties of Fiberglas," by W. M. 
Rees and R. B. Taylor, Owens-Corning Fiberglas Corp., Toledo, 


"Flame Retardant Materials for Motion Picture Applications," by 
Barton Thompson, Engineering Dept., Paramount Pictures, Inc., 
Hollywood, Calif. 

"The Filing and Cataloguing of Motion Picture Film," by C. M. 
Effinger, Twentieth Century-Fox Film Corp., Beverly Hills, Calif. 

"A Small Microphone Boom," by B. F. Ryan and E. H. Smith, 
Warner Bros. Pictures, Burbank, Calif. 

"A Survey of Photo-Template Methods," by Faurest Davis, Ansco, 
Los Angeles, Calif. 

"Power Rectifiers for Studio Lighting," by L. A. Umansky, Indus- 
trial Engineering Division, General Electric Co., Schenectady, 

8:00 p.m. Walt Disney Theater, Disney Studio, Burbank: Evening Session. 

A. M. Gundelfinger, Chairman 
Session opened with a 35-mm motion picture short. 
"The Problem of Amateur Color Photography," by R. M. Evans, 
Research Laboratories, Eastman Kodak Company, Rochester, 
N. Y. 

This was an hour and forty-five minute semi-popular demonstra- 
tion lecture on color photography. 

Adjournment of 57th Semi-Annual Technical Conference, by D. E. 
Hyndman, President of the Society. 


Position open for man or woman with experience in optical instrument 
design. Position also open for man or woman with experience in lens 
design or computing. Write for interview. Binswanger and Company, 
Optics Division, 645 Union Ave., Memphis, Tenn. 

Physicist with special training in optics for research on utilization of 
carbon arcs particularly in projection systems. Apply to Research Labo- 
ratory, National Carbon Co., Inc., P. O. Box 6087, Cleveland 1, Ohio. 

Designer and engineer experienced in optics, lighting, and microphotog- 
raphy, capable of designing microfilm reading equipment and products 
related to microfilm industry. Reply to Microstat Corporation, 18 
West 48th St., New York 19, N.Y. 

Design engineer, experienced in mechanics and optics of motion picture 
cameras, projectors, and film scanning. Give details. Reply to Mr. 
John H. Martin, Columbia Broadcasting System, Inc., 485 Madison 
Ave., New York 22, N.Y. 


Sound recording engineer, 16- or 35-mm equipment, studio or location 
work, single or double system. Free to travel. For details write J. J. K., 
354 Ninth Ave., New York 1, N.Y, 


Vol45 AUGUST, 1945 No. 2 



Report of the Subcommittee on Projector Sprocket 73 


The Motion Picture and International Enlightenment 

W. F. WANGER 76 

The Projection Life of Film R. H. TALBOT 78 

Study of Radiant Energy at Motion Picture Film Aper- 


A 16-Mm Edge-Numbering Machine L. THOMPSON 109 

Some Notes on the Duplication of 16-Mm Integral Tri- 
pack Color Films W. H. OFFENHAUSER, JR. 113 

Preliminary Report of Academy Research Council 
Committee on Rerecording Methods for 16-Mm Re- 
lease of 35-Mm Features W. C. MILLER 135 

The Calculation of Accelerations in Cam-Operated 
Pull-Down Mechanisms E. W. KELLOGG 143 

Technical News 156 

Society Announcements 158 

Copyrighted, 1945, by the Society of Motion Picture Engineers, Inc. Permission to republish 
material from the JOURNAL must be obtained in writing from the General Office of the Society. 
The Society is not responsible for statements of authors or contributors. 

Indexes to the semi-annual volumes of the JOURNAL are published in the June and December 
issues. The contents are also indexed in the Industrial Arts Index available in public libraries. 





Board of Editors 





Officers of the Society 
**President: DONALD E. HYNDMAN, 

350 Madison Ave., New York 17. 
** Past- President: HERBERT GRIFFIN, 

133 E. Santa Anita Ave., Burbank, Calif. 
** Executive Vice- President: LOREN L. RYDER, 

5451 Marathon St., Hollywood 38. 

* Engineer ing Vice-President: JOHN A. MAURER, 

37-01 31st St., Long Island City 1, N. Y. 
** Editorial Vice-President: ARTHUR C. DOWNES, 
Box 6087, Cleveland 1, Ohio. 

* Financial Vice- President: ARTHUR S. DICKINSON, 

28 West 44th St., New York 18. 
** Convention Vice- President: WILLIAM C. KUNZMANN, 

Box 6087, Cleveland 1, Ohio. 
"Secretary: E. ALLAN WILLIFORD, 

230 Park Ave , New York 17. 
*Treasurer: M. R. BOYER, 

350 Fifth Ave., New York 1. 


* FRANK E. CARLSON, Nela Park, Cleveland 12, Ohio. 
**JOHN I. CRABTREE, Kodak Park, Rochester 4, N. Y. 
**CHARLES R. DAILY, 5451 Marathon St., Hollywood 38. 

*EDWARD M. HONAN, 6601 Romaine St., Hollywood 38. 
* {CLYDE R. KEITH, 233 Broadway, New York 7. 

*G. T. LORANCE, 92 Gold St., New York 7. 
**PETER MOLE, 941 N. Sycamore Ave., Hollywood. 
*fHoLLis W. MOYSE, 6656 Santa Monica Blvd., Hollywood. 
**WILLIAM A. MUELLER, 4000 W. Olive Ave., Burbank, Calif. 

*EARL I. SPONABLE, 460 West 54th St., New York 19. 
**REEVE O. STROCK, 111 Eighth Ave., New York 11. 

*WALLACE V. WOLFE, 1016 N. Sycamore St., Hollywood. 

*Term expires December 31, 1945. fChairman, Pacific Coast Section. 
**Term expires December 31, 1946. ^Chairman, Atlantic Coast Section. 

Subscription to nonmetnbers, $8.00 per annum; to members, $5.00 per annum, included in 
their annual membership dues; single copies, $1.00. A discount on subscription or single copies 
of 15 per cent is allowed to accredited agencies. Order from the Society at address above. 
Published monthly at Easton, Pa., by the Society of Motion Picture Engineers, Inc. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 

General and Editorial Office, Hotel Pennsylvania, New York 1, N. Y. 

Entered as second-class matter January 15, 1930, at the Post Office at Easton, 

Pa., under the Act of March 3, 1879. 


Vol 45 AUGUST, 1945 No. 2 


It has long been known that the projection life of film could be in- 
creased considerably if the intermittent sprockets used were at least 
as large or larger than that required to give perfect mesh with the film 
being driven. It is on this basis that the SMPE standard sprocket 
was designed. 

When this sprocket came into use, however, it was found in some 
cases that the sprocket wore out faster than the old sprockets and the 
projector manufacturers were loathe to use them. 

It has been suggested, however, that in recent years the average 
shrinkage of film has considerably decreased and that with hardened 
and ground sprockets the difficulty of wearing of the sprockets might 
be avoided. A Subcommittee, therefore, was formed on June 19, 1941, 
to investigate the matter and bring a recommendation to the Stand- 
ards Committee. 

Approximately a year ago a series of sprockets of different diame- 
ters were made by the International Projector Company and were 
installed in special theaters where they could be observed by M. H. 
Bennett of the Warner Bros. Theatres and C. F. Horstman of the 
RKO Radio Pictures Corporation. Other sprockets made by the 
Century Projector Company have been installed in theaters in 
Rochester, N. Y., and vicinity. Before the sprockets were installed, 
careful measurements and photomicrographs of the teeth were made. 
In every case where a large-size sprocket was used the companion 
projector was fitted with a new sprocket of 0.935-in. diameter. In a 
few cases where 3 projectors were used, 2 different large-size sprockets 
were used and one 0.935-in. sprocket was used. 

The original plan was to run some of the sprockets for a year and 
some for 2 years, examining them for quietness of operation during 

* Presented May 15, 1945, at the Technical Conference in Hollywood. 



that time. At the end of the period of running the sprockets would be 
taken out of the projectors, photomicrographed for wear, and then 
would be used to determine the wearing quality of film on these 

At the present time 10 or 15 theaters have been supplied with these 
experimental sprockets and 3 sets of sprockets have been removed and 
examined. All of the projectors have been watched and the report is 
that there is no difference in noise or ease of handling the film between 
any of the sprockets. 


Runs to Break- Down with Runs to Break-Down with 

Sample Used 0.943-In. Sprocket Used 0.935-In. Sprocket 

A 1100 540 

B 843 517 

C 982 540 

D 912 491 

E 1165 630 

F 1374 573 

G 1287 566 

H 1239 530 

I 1665 590 

J 1326 603 

K 1169 594 

L 1746 800 

One set of sprockets was removed from the Riviera Theatre in 
Rochester where it had been in operation for 2 years. In this set the 
0.943-in. sprocket shows somewhat less wear than the 0.935-in. 
sprocket. These 2 sprockets were then put successively on a projec- 
tion machine and 10 different samples of film were run on both sprock- 
ets. The results are shown in Table 1. It will be observed that the 
large-size sprocket gave approximately double the film life of the 
smaller one. 

A set of 3 sprockets, 0.945-, 0.943-, and 0.935-in., was removed from 
the RKO 58th Street Theatre in New York, and a similar test run. 
These results are shown in Table 2. It will be observed that the 
0.943-in. sprocket shows somewhat better wear than 0.945-in. and 
both are considerably better than 0.935-in. sprocket. We have no ex- 
planation for the fact that the 0.943-in. sprocket appears better than 
the 0.945-in. since previous work has indicated that the reverse is 
generally ture. 

Runs to Break- 
Down with Used 
0.935-In. Sprocket 

Runs to Break- 
Down with Used 
0.943-In. Sprocket 

Runs to Break- 
Down with Used 
0.945-In Sprocket 












Runs to Break- 
Down with Used 



Although the Subcommittee had originally intended to run these 
tests for still another year, it has been suggested that the results to 
date are conclusive enough to warrant a real effort on the part of the 
Society to get theaters to use the larger-size sprockets, preferably the 
0.943-in. sprocket, in order to conserve film during the present short- 

E. K. CARVER, Chairman 







Mr. Hyndman, Ladies, and Gentlemen : 

It is a great honor to welcome the Society of Motion Picture Engi- 
neers to Hollywood, for I know of no group of men who are advancing 
the all-important cause of international understanding more effec- 
tively than you. 

It is a privilege to talk to you because you are scientists and inven- 
tors and are not bound by the past but live for the present and the 
future. You realize how communications have changed the thinking 
of the world in the past 25 years something many leaders throughout 
the world do not seem to understand. 

Your meeting really should be a part of the World Conference now 
being held in San Francisco, also and I can say this being a native of 
San Francisco the World Conference should have been held in 
Hollywood, for no medium has presented ideas more effectively to 
the people of the world than our motion pictures. And no plan 
decided upon at the San Francisco Conference can be successful un- 
less it has the support of the masses who therefore must have it pre- 
sented to them properly. The responsibility to the people does not 
end here. It goes beyond mere presentation of the basic plan. 
There must be a continuous flow of information so there will be no 
lessening of interest on their part. No peace can endure unless the 
peoples of the earth are informed and enlightened. 

We are faced with two concrete problems: How are the millions 
of people throughout the world going to receive this information and 
enlightenment, even if we put it on film, and how can we be sure the 
film inspires courageous thinking in the interest of the people and not 
controlled thinking in the interest of the few? Plato said, "Without 
the knowledge of good and evil, the use and excellence of the sciences 
will be found to have failed us." 

The answer to our first question, How the information and en- 
lightenment is to be brought to the masses, is a simple home projector 

* Presented May 14, 1945, at the opening luncheon, 57th Semi-Annual Tech- 
nical Conference in Hollywood. 

** President, Academy of Motion Picture Arts and Sciences, Hollywood. 


that can be manufactured in volume. This one device is needed 
most urgently if we are to reach the great audiences in China, India, 
Africa, and our own hinterlands, and would mean more to the world 
than any of us can realize. 

In answering the second question, I should like to quote again 
from Plato who said, in describing how each human soul before birth 
drove across heaven in the company of the gods and saw beauty, 
justice, courage, and the other virtues, that "success in life depends 
on how far, among the shadows, confusions, and distractions of earth, 
the soul retains the memory of that vision." 

Plato's story is a parable of a good education. We who have been 
dealing in popular entertainment in Hollywood have always glorified 
beauty which has become known as glamor. We have glorified jus- 
tice in all our films. Justice has always triumphed. And certainly 
we have glorified courage. As pictures have continued to be produced 
their standards have risen as public taste has improved. You have 
made it possible for us to add, first sound, which included speech and 
music, then color, and these contributions brought about great im- 
provement in all other departments of film-making. To improve 
standards in a democratic way is to improve the taste of the people so 
they will continually reject the inferior for the superior. 

The answer to the second question is to see to it that the false con- 
tent of communications will be rejected by a public fortified with 
knowledge and enlightenment as they will reject control of communi- 
cations by the selfish and greedy for their own interests. 

We are witnessing the beginning of one world. 

We can never go backward. Science is moving us too rapidly. 

The war has given the motion picture a new stature. It is now 
recognized by statesmen and educators alike as a great force for 
human enlightenment, and rightly so. For at last it has become 
evident that there is no more limitation to the use of film than to the 
printing press. 

The challenge to the makers of motion pictures is great. We must 
not limit our production to entertainment in its narrowest sense. 
We must produce documentaries. We must make educational, com- 
mercial, training, and experimental films and show them to all the 

I believe we have enough new blood among us to accept this chal- 
lenge if we have your help and enough sense to accept it. 


Summary. With the present scarcity of raw print stock, there is need for proce- 
dures which will substantially increase the useful life of the film at hand. Prints are 
rendered unserviceable by damage to the perforations, by mutilation of the edges in 
shipment, and by abrasion of the surfaces. 

The means by which each of these types of film damage can be minimized (and in 
some cases almost entirely eliminated] will be described. The adoption of only one of 
these film conservation measures should double the useful life of a print. Universal 
adoption of all these measures could make possible several hundred bookings of a print. 


There is no simple answer to the question "How many times can 
film be run through a projector?" It has been demonstrated many 
times that whereas one sample of film may become completely un- 
serviceable after relatively few projections, another sample from the 
same roll of film may still be in serviceable condition after several 
thousand projections, depending on the conditions of operation. 

It is true that there are certain qualities of the film stock itself which 
influence its wear life. Much could be written on the relation of the 
physical properties of the film (tensile strength, tear resistance, brittle- 
ness, etc.) to its durability. This relation, however, is one with which 
the film manufacturer is primarily concerned. The user of film is 
more concerned with those factors which greatly influence the wear 
life of the film he has at hand. This paper deals with some of these 

Relationship between Pitch of Sprockets, Pitch of Film, and Film 
Wear. -If motion picture film were as dimensionally stable as 
steel, there would be no problem of the relative pitches of film per- 
forations and sprocket teeth; the 2 pitches would always be equal 
and, consequently, several sprocket teeth would be in contact with 
the edges of the perforations at the same time. Actually, the pitch 

* Presented May 15, 1945, at the Technical Conference in Hollywood. 
** Eastman Kodak Company, Kodak Park, Rochester 4, N. Y. 




of the film perforations is changing more or less continually, owing 
to the loss of volatile materials and also to the dimensional changes 
which accompany variations in moisture content of the film. What 
then should be the pitch of the intermittent sprocket teeth? 

The projector manufacturer has several choices in the design of the 
intermittent sprocket. He may make the pitch of the teeth such that 
there will be almost perfect mesh between film and sprocket when the 
film is freshly processed and less and less close mesh as the film shrinks, 
or he may make the pitch of the teeth such that there will be closer 
and closer mesh as the film ages. 
He may even choose a compro- 
mise pitch so that the film fits 
the sprocket best after the film 
has shrunk a portion of the total 

In order to decide which course 
is the best to follow from the 
standpoint of film wear, it might 
be of interest to review the be- 
havior or movement of film on a 
sprocket in each of the 2 cases, 
i. e., Case I The pitch of the 
film perforations is greater than 
that of the sprocket teeth; Case 
II The pitch of the film per- 
forations is less than that of the FlG L Action of mm on a 

Sprocket teeth. sprocket when the pitch of the film 

T? 11, j.1, u t, perforations is greater than the pitch 

Fig. 1 shows the behavior or Sf the sprocket teeth. 

movement of film on a sprocket 

when the pitch of the film is greater than that of the sprocket. This 
was described before the Society in 1923 by J. G. Jones, of the East- 
man Kodak Company. 1 Note tkat the entering tooth strikes the 
edge of the perforation and shifts the film forward, the film travel- 
ing faster than the surface of the sprocket. This wedging-on of the 
film on the sprocket teeth produces a severe strain on and eventually 
a tearing of the perforations, and decreases the life of the film. This 
is the condition which exists today with the 0.935-in. diameter 
intermittent sprocket which was designed for films of much higher 
shrinkage than those in use at the present time. 

Fig. 2 shows the action of film on a sprocket when the pitch of the 



Yol 45, No. 2 

film is less than that of the sprocket. Here, the tooth that is about 
to leave the film does the driving, and the entering tooth engages the 
film without touching the edge of the perforation. The film is well 
seated at the base of the tooth when it makes contact with the driving 
face. Theoretically, this condition should cause less wear on the film 
than the previous case, in which the pitches of the film and the 
sprocket are reversed. Actually, experiments show that lower wear 
may occur on either side of perfect mesh, but, in general, so far as 
film wear is concerned, it is better to have the pitch of the film less 

than that of the sprocket rather 
than greater than that of the 
sprocket, as is the case today. 

This fact is illustrated in Fig. 
3, which shows graphically the 
increased number of projections, 
plotted along the ordinate, as the 
shrinkage* of the film, plotted 
along the abscissa, causes the 
pitch of the film to become equal 
to, or slightly less than, the pitch 
of the intermittent sprocket. 
The shrinkage of film necessary 
to give perfect mesh with a 
0.935-in. sprocket is illustrated 
by the dotted line. All points 
to the left of this line represent 
Case I, in which the pitch of the 
film is greater than that of the 
sprocket, and all points to the 
right of the dotted line represent Case II, in which the pitch of the 
film is less than that of the sprocket. 

The actual number of projections required to produce complete 
breakdown of the film at the various film shrinkages will vary widely 
with machine conditions and the manner of operation. The maxi- 
mum may occur under some conditions at several hundred to one 

* The values along the abscissa are expressed as "per cent shrinkage" of the 
film, for convenience. Since it would be impossible to obtain shrinkages of such 
great range with the present motion picture films, many of the samples were per- 
forated less than standard pitch on specially designed equipment in order to simu- 
late films having these different shrinkages. 

FIG. 2. Action of film on a 
sprocket when the pitch of the film 
perforations is less than the pitch of 
the sprocket teeth. 

Aug., 1945 



thousand passages through the projector; under still better condi- 
tions, it may occur at several thousand passages. Cases have been 
noted in which the maximum occurred at 22,000 passages. These 
figures were obtained by projecting short loops of film continuously 
on a simplified machine under carefully controlled conditions. The 
number of projections therefore are probably higher than can be ob- 
tained in practice because of the carefully controlled conditions under 
which such tests must be run. The main point is that there is a 

FIG. 3. Relationship between the pitch of the film 
and the number of projections necessary to produce film 
breakdown, using an 0.935-in. intermittent sprocket. 

maximum and that this maximum occurs on a 0.935-in. sprocket at a 
shrinkage of film far greater than exists in any present-day motion 
picture positive film. (The amount of shrinkage of most motion pic- 
ture positive film lies between 0.0 and 0.6 per cent during its normal 
projection life.) 

Theoretically, the film is in perfect mesh with the 0.935-in. sprocket 
when it has a shrinkage of 1.2 per cent, and one might expect that 
the maximum should occur at this point. Frequently, it does occur 
close to this point, as indicated in Fig. 3. However, it occurs often 
at a point well beyond the point of perfect mesh, as illustrated by Fig. 
4. The reason for the failure of the maximum passages to occur at 



Vol 45, No. 2 

the point of perfect mesh has not been clearly established. The fol- 
lowing is offered as a possible explanation : In Case I, in which the 
pitch of the film is greater than that of the sprocket, the film wedges 
on to the sprocket tooth, the film traveling faster than the circumfer- 
ence of the sprocket. Consequently, all of the driving action is ac- 
complished by the sprocket teeth. In Case II, in which the pitch of 
the film is less than that of the sprocket, the circumference of the 
sprocket travels faster than the surface of the film. In this case, a 
considerable portion of the driving action is accomplished by contact 

1.2 1.6 2.0 


FIG. 4. Same as Fig. 3 with the exception that the maxi- 
mum number of projections does not occur at or near the 
point of perfect mesh of film and sprocket. 

along the circumference of the sprocket as well as by the sprocket 
teeth, thus causing the number of passages to increase to a certain 
point as the pitch of the film becomes less than that of the sprocket. 

Likewise, the failure of the maximum number of projections to oc- 
cur at the calculated point of perfect mesh can be explained by the 
stretching of the film at the point of impact. If the film stretches 
upon impact of the sprocket tooth, and this appears to be a reasonable 
assumption, the effective pitch is greater than the calculated or meas- 
ured pitch. Therefore, to obtain the optimum projections, the pitch 
of the film should be less than that required to give perfect mesh. 

There is another and perhaps more satisfactory manner in which 

Aug., 1945 



data of this type can be presented graphically. Since it is not the 
actual pitch of the sprocket and of the film, but only the difference 
between them which matters, we may express this difference graphi- 
cally along the abscissa as the per cent deviation of the film from the 
pitch of the sprocket, as in Fig. 5. Zero on the abscissa, therefore, 
represents a perfect fit of the film on the intermittent sprocket in 
question, and, as before, all points to the left of the line represent Case 
I, in which the pitch of the film is greater than that of the sprocket, and 

H.O +.5 -5 -1.0 

FIG. 5. Relationship between the per cent deviation 
of the film from the sprocket pitch and the number of pro- 
jections that produce film breakdown. 

all points to the right of this line represent Case II, in which the pitch 
of the film is less than that of the sprocket. Thus, films of any pitch 
whatsoever may be run on projectors with intermittent sprockets of 
different diameter and, if other factors are controlled, the results will 
still be comparable. For example, the data presented by D. R. 
White 2 at the 1943 Spring Meeting, in which films of nearly stafidard 
pitch were run on a Series of Different diameter intermittent sprockets, 
can be presented graphically in this manner and compared with the 
results just given. 

These principles have been recognized by film manufacturers and 
projector manufacturers alike. The approval of the 0.945-in. diame- 



Vol 45, No. 2 

ter intermittent sprocket and its adoption in 1930 as an American 
Standard came as a result of this general agreement. Why, then, has 
the 0.945-in. sprocket not been in general use in this country?* 

The answer is that the projector manufacturers have been con- 
cerned over the possibility of increased wear of the sprockets, their 
fears being based on the theory then held. It is quite obvious that 
if we consider the film as a steel tape (Fig. 1), the wedging action of 


FIG. 6. Wear on 0.943-in. sprocket after 20 months' use in Riviera Theatre, 
Rochester, New York. 

the film onto the entering tooth should wear away the face of the 
tooth uniformly. When, however, the pitches of the film and the 
sprocket are equal, or the pitch of the film is less than that of the 

* As early as 1933, larger-diameter intermittent sprockets were in use in several 
European projectors. In 1933, intermittent sprockets used on Ernemann pro- 
jectors in France and Germany measured 0.9452 in. in diameter, and those on the 
Kalle projectors in England, 0.941 in. in diameter. The Phillips projector, which 
was described before the Society by T. W. M. Schaffers, 3 is equipped with 0.945- 
in. diameter intermittent sprockets. When conditions permit, a survey of the 
results obtained from the use of these larger-diameter sprockets in Europe should 
be of value. 

Aug., 1945 



sprocket, as in Fig. 2, the entire wear should occur at the point of 
contact or at the base of the tooth. The wearing away of the base 
of the tooth should give it a hooked shape which would result in rapid 
film wear. 

The principal objective of the theater tests, as outlined by the Sub- 
committee on Projector Sprocket -Design, was the determination of 
the comparative wear of the 0.935-in. and the oversized intermittent 

FIG. 7. 

Wear on 0.935-in. sprocket after 20 months' use in Riviera Theatre, 
Rochester, New York. 

sprockets. Fortunately, our former theory in regard to sprocket wear 
was somewhat in error. In every theater in which these oversized 
sprockets have been compared with the 0.935-in. ones, the latter have 
%hown the more wear. In Figs. 6 and 7 are shown the various degrees 
of wear on each of 4 teeth from a set of sprockets removed after 20 
months' operation in the Riviera Theatre in Rochester, N. Y. Fig. 6 
shows the wear on the 0.943-in. sprocket. The wear on this larger- 
diameter sprocket is predominantly at the base, as the theory sug- 
gested, but it is slight and not sharply defined. There is no evidence 
of a hook. The wear on the corresponding 0.935-in. sprocket is 


Vol 45, No. 2 

shown in Fig. 7. Note that in the case of this sprocket, the film has 
cut deep grooves into the face of the teeth. The striking thing about 
these grooves is that they vary in depth and in the distance from the 
base of each of the 4 teeth. 

A similar comparison is given in Fig. 8 which shows the various de- 
grees of wear on each of 4 teeth of a set of 3 sprockets removed from 
the RKO 58th Street Theatre, New York City, after 1552 hrs projec- 

FIG. 8. 

Comparative wear on 3 sprockets used in RKO 58th Street Theatre, 
New York, for 1552 hr. 

tion time for each sprocket. Again, the wear on the 0.935-in. sprocket 
is greater than that on the oversized sprockets. This difference in 
type and extent of wear is more apparent when the sprocket teetfc 
are examined with the aid of a binocular microscope than in the pho- 
tomicrograph. Obviously, the analogy of the sprocket and the 
steel tape to the case of the sprocket and the film is not complete. It 
is hoped that in the near future there will be more concrete evidence 
of the nature of the action of film on intermittent sprockets during 
the pull-down cycle. 

Aug., 1945 



The foregoing remarks relate to the difference in the nature of the 
wear on the sprocket teeth. The increased extent of wear on the 
0.935-in. intermittent sprocket as compared to that on the oversized 
sprocket is explainable. In the case of the former, the additional 
force necessary to thrust the film ahead of the rotational speed of the 
sprocket manifests itself in increased wear of both film and sprocket 
teeth. When the pitches of film and sprocket are equal, the driving 
force is spread over 2, or even 3 teeth simultaneously resulting in 
decreased wear on both film and sprocket. This condition should 
hold true even when the 2 pitches are nearly equal, as a result of the 
local stretching of the film upon impact. 




FIG. 9. Impact marks of an intermittent-sprocket 
tooth on a film perforation. Case I Correct Alignment; 
Case II Incorrect Alignment. 

It is therefore gratifying that by the use of increased diameter inter- 
mittent sprockets substantially increased film life can be achieved 
with no increase in sprocket wear. 

Relationship Between the Alignment of the Film on the Inter- 
mittent Sprocket and Film Wear. Fig. 9 is a drawing of the marks 
made on the standard positive-type perforation by the intermittent 
sprocket on one passage through a projector. It is impossible to 
align the film on the intermittent sprocket so that the latter will 
strike .the film in the center of each row of perforations. This is 
because the transverse pitch of the sprocket is less than that of the 
film. If the film is centered on the intermittent sprocket on one 



Vol 45, No. 2 

row of perforations, the sprocket will ride very near the inside corner 
of the other row of perforations, as in Case II, and wear or rupture 
will then almost invariably occur at the point nearest the sprocket 
tooth, or at A '. It is believed that the best alignment with the pres- 
ent intermittent sprocket occurs when the sprocket is centered as 
nearly as possible on each row of perforations, or so that distances A 
and A' are equal, as in Case I. Even under this condition, the pre- 
ponderance of wear occurs in the corners at A and A ' and occasionally 
at the outer corners also. Only rarely does it occur directly under the 
point of impact, in the case of the 0.935-in. intermittent. This is 
because the perforation is weakest at the points of curvature, the 


FIG. 10. Correct alignment of film with standard 
positive-type perforations on an intermittent sprocket 
(actual photograph). 

wedging-on action causes a strain throughout this entire area, and 
rupture occurs at the weakest point. 

Correct alignment of the film on the intermittent sprocket in the 
case of the standard positive-type perforations is illustrated in Fig. 
10.* Note that the distance from the inside edge of the left-hand 

* The film was allowed to make one passage through the projector, thereby 
leaving the mark of the intermittent on the perforation. The film was then 
placed in a Recordak Viewer, a projection made of the marked perforations onto 
graph paper, and the superimposed image of the perforation on the graph paper 
photographed. In practice, millimeter paper is used for mor& precise alignment. 
In these photographs, l / 2 -cm grids were used to facilitate reproduction. 

Aug., 1945 



perforation to the point of impact of the intermittent-sprocket tooth 
is nearly 3 divisions of the grid, and that the corresponding distance 
on the opposite row of perforations is again nearly 3 divisions. In 
other words, distances A and A ' are equal. Perfect alignment can 
rarely be obtained by predetermination from measurements made on 
the projector, i. e., by mechanical setup. It can best be accom- 
plished by allowing the intermittent sprocket to mark the perforations, 
and then, by projection of these markings on graph paper, making the 
necessary adjustment in alignment. 


15 ?0 


FIG. 11. Relation between the tension in the gate and 
the number of projections that produce film breakdown. 

Although there are no data on the subject available at this time, 
experiments have indicated that double the number of projections 
could be obtained from films which are exactly centered on the inter- 
mittent sprocket compared with those which are only slightly mis- 

Relation Between the Tension on the Film in the Projector Gate 
and Film Wear. The relation between the tension of the film in 
the gate and the number of projections necessary to produce com- 
plete breakdown is shown in Fig. 11. It will be noted that on this 
particular machine the number of passages vary from about 300 at 

90 R. H. TALBOT Vol 45, No. 2 

a tension of 30 oz* to 3000 at a tension of 8 oz. At tensions lower 
than 8 oz, the samples were not run to breakdown, but were re- 
moved after 3000 passages and examined for wear. 

On the projector on which these tests were run we could see no 
difference in steadiness of the screen image with tensions greater 
than about 4 oz, but below this point there was definite unsteadi- 

A survey of a number of theater projectors in our district showed 
gate tensions ranging from about 20 oz to about 10 oz. From the 
curve we see that these 2 extremes in tension would produce a fivefold 
change in the wear life of film. 

This relationship between the number of projections and the gate 
tension represents but one series of tests on one type of projector. On 
this projector, the gate tension may be changed by means of a thumb- 
screw which alters the loading of the central coil springs which, in 
turn, controls the pressure on the center tension pads. 

There is some evidence to indicate that whereas the over-all effect 
of tension on most projectors may be the same as that just given, it is 
much more difficult to determine. For instance, in many bf the older 
types of projectors the gate tension is controlled by individual canti- 
lever springs, the adjustment of which can be made only by hand 
manipulation, i. e., by bending the springs. Moreover, on many 
types of projectors in which the film is not positively edge-guided 
through the gate, a change in gate tension produces a change in align- 
ment which in itself alters the result. 

The answer, then, to the question "How many times can film be 
run through a projector?" appears to be "It will run as many times 
as we wish to make it run." If the pitch of the present intermittent 
sprocket is altered so as to avoid the tearing action of the sprocket 
teeth against the film perforations, and if the questions of alignment 
and tension can be solved in cooperation with the projectionist, it is 
believed that most films are capable of delivering many times the num- 
ber of projections they will ever be called upon to make. 

* The tension in the gate can be measured easily by inserting a short strip of 
film in the gate and withdrawing it upward through the gate by the steady pull of 
a hand scale, or if preferred, by other mechanical means, so that the force required 
to move the strip in the gate can be measured. On a theater projector, the upper 
magazine must be removed to facilitate withdrawing of the test strip, or a special 
jig must be used which allows the tension to be measured with the magazine in 



All too frequently a print is received by the exchange directly from 
a theater with torn edges, or in some instances with one row of per- 
forations entirely missing from some sections. How can this be? 
In some cases the print has been so badly damaged that projection in 
this condition would have been impossible. Yet the theater had 
made no complaint nor had they asked for another print. We must 
assume, therefore, that the print operated satisfactorily. Likewise, 
the reverse is true. Prints have been received by the theater di- 
rectly from the exchange with the edges broken to the extent that 
several sections (2 to 3 ft in length) had to be removed before the roll 
could be projected again, and yet this print had been inspected care- 
fully at the exchange and the inspector's seal affixed thereto. There 
is a preponderance of evidence to the effect that most of the damage 
occurred between the theater and the exchange. 

The writer is aware that a certain amount of edge damage is caused 
by the film coming into contact with the sharp surfaces of defective 
reels a condition brought about largely by the scarcity of new metal 
reels at this time. He is also aware that in certain rare cases, which 
each one of us can recall, the edges of the print have been damaged as 
a result of what we may call an accident, i. e., the reel of film may 
have been dropped, the metal shipping case may have been dented or 
crushed, or the film may have jumped a sprocket at a splice. But 
let us not, in recalling some of these instances, assume that the 
major amount of edge damage comes from these causes. It is far too 

There is another observation which is very pertinent to this matter 
of edge damage, i. e., the damage is confined largely to prints which 
have been in use long enough to become oily. It is very difficult to 
wind a smooth firm roll on a hand rewind when reels, which are 
wider than the film and often niisshapen are used. Uneven wind- 
ing leaves the edges of certain convolutions protruding farther than 
those immediately adjacent. If the roll is not wound too tightly, 
and if the film is not oily, these protruding edges will slip back into 
place by pressure from the reel flanges. If the film is oily, however, 
a firm bond is formed between successive layers within the roll so 
that the protruding edges of an unevenly wound roll cannot slip back 
into place. When pressure is brought against these edges by the reel 
flanges during shipment, or even by careless handling, they will be 
damaged and, in some cases, even broken off completely, as shown 



Vol 45, No. 2 


FIG. 12. An example 
of edge damage on posi- 
tive motion picture film. 

in Fig. 12. Basically, the solution of this 
problem would be to ensure that all rolls are 
evenly wound and that no oil is allowed to get 
on the film. This would not be very practi- 
cal, for both of these conditions will probably 
prevail for some time to come. 

This problem of edge damage can be at- 
tacked from 2 angles: 

(1) By providing protection for those edges of film 
which protrude; 

(2} By surface-treating the film in such a manner 
that the protruding edges of film, even when oily, will 
telescope back into the roll without damage. 

The first of these 2 methods, protection 
of the protruding edges, was discussed in a 
paper read at a recent meeting of the 
Society. 4 This proposed solution concerns 
the use of a white plastic band to replace the 
paper wrapper on exchange reels. The band 
which was described was approximately 0.040 
in. thick and 3 /i 6 in. wider than the film. The 
greater width of the band, together with its 
stiffness, prevents to a large degree the bend 
ing over of the protruding edges of the film by 
the reel flanges. 

A test was made to demonstrate the effec- 
tiveness of this type of band for the preven- 
tion of edge damage. Three 2000-ft rolls of 
processed positive motion picture film were 
spotted with oil throughout their entire 
length. Each roll was then projected 10 
times in order to "bake" the oil into the 
film. The rolls were then mounted on 3 
identical exchange reels, using a constant- 
tension winding mechanism. In the mount- 
ing of these rolls the film was forced first 
against one flange and then against the op- 
posite flange in order to produce protruding 
edges. Around one roll was placed the con- 
ventional paper wrapper or band. Around 

Aug., 1945 



another roll was placed the white plastic band. Around the third 
roll was placed a similar stiff band which instead of being plastic, 
was made of black pressboard. These reels of film were then placed 
in a 3-reel shipping case and given a handling test to simulate the 
abuse many cases of film undergo in shipment. The positions of 
the reels in the case were changed from time to time. 

At the end of the test, the bands were removed and the reel flanges 
bent back so that the extent of edge damage could be seen. The 
result of the test is shown in Fig. 13. It can be seen that there is 

FIG. 13. Comparative edge condition of film protected by stiff bands 
(not shown in picture) with edge condition of film wrapped with customary 
paper band. 

some edge damage on the 2 rolls protected by the stiff bands, prob- 
ably as a result of the severity of the treatment, but the edge damage 
on the roll covered by the paper wrapper is considerably greater. It 
is believed that by this simple expedient a large percentage of the edge 
damage to prints can be prevented. 

The second method of attack, surface treatment, has been the sub- 
ject of numerous investigations, 5 over a period of many years, often 
leading to patented processes. The principal objection to most of 
the existing processes lies in their inability to satisfy all the require- 
ments of the ideal surface treatment at the same time. It is believed 
that an ideal surface treatment should accomplish the following: 



Vol 45, No. 2 

(1) Provide a smooth coating of high gloss which will minimize the mottle on 
the screen from oil spots on the film. 

(2) Give projection runs equal to or surpassing film having a liberal applica- 
tion of paraffin wax along the perforated area. 

(5) Minimize or eliminate the edge damage previously mentioned by provid- 
ing a slippery surface even when the print is oily. The rating of films for their 
ability to telescope when oily will be referred to hereafter as their "oily slip." 

(4) Provide protection from surface abrasion. 









Candelilla Japan 


Carnauba Curicury 

Synthetic and Trade Name 

Albacer 1163 

Dioctadecyl carbonate 

Pentaerythritol diacetal 

Albasol BB 


Pentaerythritol di-N-butyral 


Glyceryl stearate 

Pentaerythritol di stearate 

Ammonium stearate 

Glyco dilaurate 

Pentaerythritol mono stearate 

Amorwax 1200 A 


Pentaerythritol tetra stearate 


Glycowax B 430 



Glycowax A 1639 

Propylene stearate 

Barnsdall special 


Rezowax A 

Betanol 107 


Rezowax B 

Betanol 114 

I.G. Wax E 

Santowax M 

Betanol 152 

I.G. Wax O. P. 

Santowax MH 


Johnson's WM 119 

Santowax O 


Johnson's WM 169C 

Santowax P 


Laurie ethanolamide 

Santowax R 


Old English Impervium 

Span 40 

Cetyl alcohol 



Diglyco stearate 

Pentawax 217 

Witco Hamp No. 70 

Since it is believed that no commercially available surface treat- 
ment satisfies all these requirements, a study is now being made of 
the problem. The data to be presented and the conclusions to be 
drawn are in the nature of a progress report. The scope of the prob- 
lem and its complexity preclude a final answer at this time. 

A large number of waxes were examined, as shown in Table 1, for 
their solubility characteristics and their general appearance when ap- 
plied to the surfaces of film in any reasonable concentration. Various 
concentrations of these waxes in carbon tetrachloride were applied 

Aug., 1945 



to the entire emulsion surface of positive films by the plush-wick 
method on a machine designed for this purpose. 

From the preliminary survey of this large number of samples only 
the waxes shown in Table 2 were considered worthy of further testing. 


Waxes judged most suitable for surface lubrication of motion picture positive film. 
For purposes of comparison, the samples were given an alphabetical rating for oily 
slip and scratch resistance "A" representing the ideal. 




Pentawax 217 

Glyco dilaurate 

Johnson's WM 169C 



Pentaerythritol tetra stearate 


Santowax R 




Betanol 107 

Betanol 114 

Betanol 152 


Pentaerythritol diacetal 

Pentaerythritol di-N-butyral 

Dioctadecyl carbonate 

Johnson's WM 169C (Repeat 

Johnson's WM (Both Sides) 

Eastman Edge Wax (Perfor- 
ated Area Only) 

Max. Useful 
Per Cent 
Cone, in 


Wear Life as 
Per Cent of 














































Oily Slip 




































B + 








One thousand-ft rolls of film were coated with each wax from the con- 
centrations in carbon tetrachloride indicated, and examined for ap- 
pearance, wearing quality, oily slip, and scratch resistance. Note 
how low the concentrations of wax must be to give a surface coating 
of good appearance. Most wax applications in these low concentra- 
tions give only approximately one-half the wear life of edge-waxed 
film. In the oily-slip test, at least 900 ft of each sample roll was 

96 R. H. TALBOT Vol 45, No. 2 

liberally spotted with machine oil throughout its entire length and 
then projected 10 times on a representative theater projector to bake 
the oil into the film. Apparently, many waxes, together with oil, 
yield a greaselike composition which causes the layers of film to ad- 
here together, since in many cases the lubricated films have a lower 
rating on the oily-slip test than untreated films. 

Of the waxes examined thus far, only one seemed to be outstanding 
in respect to the telescoping of an oily roll Johnson's Industrial 
Wax WM -169-B. For some reason, the waxes used in this mixture 
give a smooth, glossy coating which imparts a slip to the film layers 
even when the film is oily. In the first test, the wearing property 
was not satisfactory. In a repeat test, it was much better, and when 
applied to both sides of the film, it was better than that of the edge- 
waxed sample. It is admitted that the application of wax to both 
sides of film may present other problems such as difficulty in splic- 
ing, etc. 


Any simple wax application depends for its effectiveness on the 
principle that the treated surfaces will be more resistant to abrasion 
than those not similarly treated. Whereas this is true in some cases, 
as seen in Table 2, the fact remains that no practical film surface has 
been found which will resist abrasion indefinitely. Therefore, when 
these treated surfaces become abraded, they present the same problem 
as do any other scratched films. 

We may distinguish, then, between the resistance offered to abra- 
sion by a simple lubricant and the protection of the film surfaces from 
abrasion by a thicker coating, such as a lacquer. In addition, it is 
necessary not only to have the lacquer coating of such a thickness 
that it will carry the abrasion which normally is borne by the film sur- 
faces, but, to be really effective, it must be removable and renewable. 
Thus, at any time in the life of the film, the entire external surfaces of 
the film can be renewed. Only in this way can "new print" quality 
be maintained. 

At a meeting of the SMPE in Hollywood in 1940, a method for the 
scratch protection of motion picture film which fulfilled these require- 
ments was presented. 6 The novelty of the method lay not in the 
fact that it was a smooth, glossy protective coating for either or both 
surfaces of nitrate and safety films, but in the 'fact that the protective 
coating was removable in dilute alkali (or developer) and thus could 


be replaced from time to time, if necessary. The fact that new print 
quality, in regard to oil mottle and all normal abrasion, could be 
maintained throughout the life of a print was demonstrated at that 
time. A print was placed at our disposal by J. M. Nickolaus of 
Metro-Goldwyn-Mayer Studios. After treatment, the print was 
put into service through the Buffalo exchange of MGM. A portion 
of the print was shown here after 35 bookings or an estimated 164 runs. 
The conclusions reached at that time were that : 

(1) The lacquered section was noticeably more free from oil mottle and 
scratches than the unlacquered section; 

(2) The section which had its original lacquer removed and replaced by a 
fresh coating prior to its showing here was judged to have "new print" quality. 

The Ace Laboratory, Brooklyn, has used the lacquer for several 
years on their negatives as an insurance against abrasion. The Ace 
Laboratory has also used the lacquer on both surfaces of their color 
film, mainly for the prevention of color distortions arising from the ef- 
fect of oil on the surface of the film. 

The process has never been used extensively for positive film. This 
may be because in the past it was easier to replace the occasional 
severely scratched print than to insure every print against abrasion. 

Furthermore, since the beginning of the war, the materials used in 
this lacquer have been on high priority. On account of the existing 
print shortage, every effort is now being made to make the lacquer 
available, even for use on prints. 

If the original lacquer, as announced in 1940, is compared with the 
simple waxes, it would be rated as follows : 

(a) Good in appearance and oil-mottle prevention. 

(6) Unsatisfactory for wearing property without additional lubrication. 

(c) Poor for oily slip. 

(d) Fair to poor for scratch resistance. 

(e) Very good for scratch protection. 

Recently all of the available waxes were tried separately and, in 
some cases, in combinations, as an integral part of the lacquer. Sev- 
eral combinations were found to be an improvement over either the 
wax alone or the lacquer alone. One combination was found to be ex- 
ceptionally good for scratch resistance, good for oily slip, but not 
satisfactory for wearing quality. If, however, the print is edge-waxed 
in addition to the lacquer-wax coating, the treatment seems to satisfy 
all the requirements previously given for an ideal surface treatment. 

98 R. H. TALBOT Vol 45, No. 2 

The work will be continued. Perhaps something more simple and 
effective will be discovered. Extensive trade tests will have to be 
made on the most promising developments. 


It has been shown that an increase in the wear life of film will result 
if the diameter of the intermittent sprocket is increased from 0.935 
in. to 0.943 in. This fact has been demonstrated many times by lab- 
oratory tests. The results in the laboratory have been verified by 
trade tests in which it has been shown that the 0.943-in. intermittent 
sprocket will give from 2 to 3 times the number of projections, before 
breakdown of the film, as the 0.935-in. sprocket after each sprocket had 
been in theater use for over 1500 hr. 

Laboratory tests indicate that a decrease in gate tension would 
increase the wear life of film severalfold. 

The centering of the perforations on the intermittent sprocket has 
a considerable bearing on the wear life of film. 

Mention has been made of the necessity for the reduction of the 
damage to the edges of prints. Two methods by which this damage 
can be minimized have been described. 

Abrasion of the surface of film can be reduced by the application of 
wax to the entire surface. Surface abrasion can be eliminated almost 
completely by the application of a renewable lacquer. 


The writer wishes to express his sincere appreciation to Dr. E. K. 
Carver for his many helpful suggestions and for continued guidance in 
the preparation of the paper, and to the various members of the film- 
testing departments of the Eastman Kodak Company for their con- 


1 JONES, J. G.: "Film Sprocket Design," Trans. Soc. Mot. Pict. Eng., 17 (Oct. 
1923), p. 55. 

2 WHITE, D. R., AND DEMoos, C.: "A Note on the Projection Life of Film," 
/. Soc. Mot. Pict. Eng., 41, 4 (Oct., 1943), p. 297. 

3 SCHAFFERS, T. W. M. : "A New 35-Mm Projector with a New Light Source," 
J. Soc. Mot. Pict. Eng., 44, 3 (Mar., 1945), p. 203. 

4 KULKA, T.: "A Greatly Improved Reel Band," presented Oct. 16, 1944, at 
the SMPE Technical Conference in New York. 

6 CRABTREE, J. I., SANDVIK, O., AND IVES, C. E.: "The Surface Treatment of 
Sound Film," J. Soc. Mot. Pict. Eng., 14, 3 (Mar., 1930) p. 275; also CRABTREE, 


J. I., AND IVES, C. E., "The Lubrication of Motion Picture Film," Internal. Pro- 
jectionist, 3 (July, 1932), p. 7. 

TALBOT, R. H.: "A New Treatment for the Prevention of Film Abrasion 
and Oil Mottle," /. Soc. Mot. Pict. Eng., 36, 2 (Feb., 1941), p. 191. 


MR. HERBERT GRIFFIN : The paper is now open for discussion. 

MR. M. S. LESHING: A few years ago the Eastman people advocated a protec- 
tive lacquer treatment of negative film right after processing. After a period of 
time the question of the lacquer died. What happened to it? 

MR. TALBOT to MR. D. E. HYNDMAN: Don, would you like to answer that 

MR. HYNDMAN: To my knowledge the Eastman Kodak Company, as a policy, 
did advocate applying a protective lacquer to either processed positive or negative 
film, but the use of it was never discouraged. 

MR. LESHING: The reason, I think, that the Kodak Company planned using 
a lacquer is shown by the fact that Mr. Capstaff was, as he told us, providing the 
developing machine in the Kodak Research Laboratories with a special cabinet 
for applying the lacquer to the negatives. 

MR. HYNDMAN: The only answer I can give would be my personal opinion. 
The Eastman Kodak Company has sold and delivered lacquer to anyone desiring 
it. It was and is still used mainly by the Ace Film Laboratories in Brooklyn. 
New York. It has used lacquer for a period of years. I have inspected, person- 
ally, negatives on which lacquer has been applied. After 475 release prints have 
been struck from the negative, the lacquer was removed. The concensus of 
opinion was that the negative looked practically as good as before printing. For 
this reason Ace Film Laboratories, Inc., continued to use the lacquer. We 
would sell the lacquer to anyone who wishes it, but we have not tried to thrust its 
use on any concern. It is available. 

MR. TALBOT: As a service of the Kodak Company we will provide specific 
recommendations on equipment to apply the lacquer. 

MR. GRIFFIN: Any more discussion, please? 

MR. PAUL ALLEN: It seems to me that the industry is missing an opportunity. 
If it would purchase 30,000 sprockets at about $150,000, it would save about 
half of the film cost. There are the facts that should be presented. 

MR. GRIFFIN: As a representative of the company mentioned, the Inter- 
national Projector Corp., probably one of the largest manufacturers of sprockets 
in the world, I have been most interested in this subject for many, many years 
and am heartily in accord with what this paper has shown. As a matter of fact 
I presented the same type of paper in the past which I read before the Society in 
1933 and which was published in the January 1934 issue of our JOURNAL. How- 
ever, on the basis of that I issued a preemptory order to our Engineering Depart- 
ment to change all intermittent sprockets from 0.935 in. to 0.945 in. diameter 
which we promptly did and probably shipped several thousands. Shortly there- 
after we were called upon to replace sprockets for the reason that there was a 
noise regardless of what the previous speaker said. A noise would show up from 
a slight undercut of the pull-down tooth. We replaced all of the 0.945 sprockets. 
I agree that in those days the shrinkage problem was more acute than it is today. 

100 R. H. TALBOT Vol 45, No. 2 

However, that situation did exist with the result that we were called upon to 
replace thousands of sprockets. Our company is in business to make money 
and cannot continue to replace sprockets. The projectionists just would not use 
them. I agree with the gentleman who said it would be a wise plan on the part of 
the film manufacturers to supply the sprockets themselves, but I don't think 
that would work either, as they would have the same problem and the same re- 
sults. Anybody like to rebut that? 

MR. TALBOT : May I ask you one question ? Are you sure that those sprockets 
used at that time were hardened steel or were they soft steel? 

MR. GRIFFIN: They were hardened and ground steel. The radii were ground 
also on all of the teeth and still are. 

MR. TALBOT: The only thing that I can say, more or less in rebuttal, would 
be that if these sprockets have been in use for approximately one year and still 
give double the life of film, it would pay the exchanges to replace the sprockets 
today if they had to. I mean, if after a year they are still getting better runs, 
even if it were true at the end of a year the sprockets began to wear and had to be 
replaced, it would still seem worth while for the exchanges to do this rather than 
have to buy so much film. 

MR. GRIFFIN: With that I heartily agree. Nevertheless we had to shoulder 
the burden and would not want to have to do it again. I still want to recommend 
a 0.943-in. sprocket. 

Any further discussion? 

MR. E. J. DENISON: Regarding edge damage to film, mentioned in Mr. 
Talbot's paper, it has been my experience over a period of years covering film 
exchanges that a great deal of the edge damage is actually done during rewinding 
of the film in the exchanges and not by projectionists or projection machines. 
If one will visit the average film exchange today he will find that rewinds are 
badly out of alignment, with the result that the film traveling from the free- 
running reel to the rewinding reel is drawn across the side of the rewinding reel, 
resulting in the film being broken through from the outer edge to the perforations. 
Further, the out-of-line setup of the rewinds results in some convolutions of the 
film protruding from the side of the roll. When one reel is stacked on top of 
another during handling by the film shipper or truck driver, these convolutions 
which protrude from the roll are broken down by weight, resulting, of course, in 
considerable edge damage to the film. 

The matter of edge damage to film has been discussed before the National 
Film Carriers Association for several years past, with demonstrations as to the 
cause of this particular type of damage, as well as the SMPE, without result. 
The National Film Carriers Association has taken serious steps in the past to 
prevent undue rough handling by their drivers, but to my knowledge they have 
had little success. 

From time to time the .idea has been advanced that some of this edge damage 
is due to poor quality of film which, of course, is not true. A number of years 
ago I made a slow-motion picture for Paramount showing damage to film due to 
improper splicing. This picture demonstrated the necessity of proper splicing 
in order to prevent damage to the film at the point of splicing. Today several 
of the distributors have their exchanges equipped with a modern and efficient 
splicing device, with the result that these distributors have very little film dam- 


aged due to splicing. Befpre making the picture demonstrating results of bad 
splicing, I gathered film scraps from some 35 exchanges throughout the United 
States, and analyzed and catalogued each particular type of damage due to im- 
proper splicing. The result of this analysis was the slow-motion film mentioned. 

The point I am trying to make is this: I believe that in order to overcome 
excessive edge damage to film, or any other type of damage, it is necessary to 
visit a sufficient number of exchanges throughout the country, gather samples of 
film damage, analyze and catalogue them. Any particular type of damage to 
film can quickly be catalogued by an expert and only then can proper steps be 
taken to correct this evil. In my opinion, testing or analyzing film damage cannot 
be satisfactorily arrived at in a laboratory. The life of motion picture film has 
been greatly increased in the past years due to intelligent investigation and analy- 
sis of causes. There is still considerable research to be done to bring about the 
maximum life of positive prints and most of these causes will be found in the field, 
mainly in the film exchanges. 

With the advent of sound it was necessary to keep projector heads in a much 
better mechanical condition than in the silent days. There have been many 
attempts to correct sprocket pitch, take-up tension, use of bad reels, etc., all of 
which are contributing factors to various types of film damage. 

Positive film is handled only in exchanges and theaters, consequently it is my 
belief that a thorough survey of these 2 branches of the industry should result in 
the establishment of the necessary standards for the proper handling of film. 

First, the investigation should be made in the exchanges where, it will be found, 
a great deal of the damage to film occurs. There should be designed a good re- 
wind and a blueprint for the installation of these rewinds. Also it is considered 
practical to install between the rewinds a guide for keeping the convolutions of the 
roll of film smooth in its travel from the free-running reel to the rewinding reel. 
Good reels should always be used in connection with the inspection and rewinding 
of film in exchanges. All devices used in the exchange in connection with inspec- 
tion and handling of film should be engineered into the general setup. 

In the past there has been a number of treatments for positive film tested. 
Some were found beneficial, others showed no improvement in the life of the film. 
Today it is definitely known that there are certain treatments for positive film 
that actually retards scratching, keeps the film pliable, retards oil absorption, 
and, in general, while not necessarily prolonging the life of the film by such a 
treatment, will keep the film in a new condition for a longer period of time. 

It is this speaker's thought that it would not be amiss to appoint a committee of 
experts to make a thorough survey on the care and handling of film. 

MR. GRIFFIN: I hope the Editorial Board will find it possible to promptly 
publish this paper. It will be most interesting to all projection manufacturers 
including my own company. Thank you, Mr. Talbot. 




Summary. The results of a study of the quantity and quality of radiant energy 
incident at the center of the film aperture with various carbon arc motion picture pro- 
jection systems are reported. The effect on the spectral quality and quantity of the 
radiant energy of various filters used primarily to remove nonvisible energy is also 

In motion picture projection, radiant energy from a light source is 
concentrated on the film at the projector aperture and transmitted by 
the projection lens to form the picture image on the projection screen. 
Previous publications in the JOURNAL have given pertinent data con- 
cerning the carbon arcs and lamps and optical systems commonly em- 
ployed for this purpose. The useful portion of the total radiant energy 
of the projected beam is that of wavelengths visible to the human eye. 
Information has been published relative to the spectral energy dis- 
tribution throughout the visible wavelength region both for the light 
from the bare arcs 1 and for the light falling on the projection screen. 2 
This visible radiation is accompanied by a certain amount of radiant 
energy of wavelengths outside the visible region, particularly in the 
infrared region. The heating effect of this energy is of interest and 
importance in connection with the development of improved arcs and 
optics to concentrate higher light intensities on the aperture. At- 
tention has been focused on this heating effect by the recent work of 
Carver 3 and his associates wherein it has been shown that under some 
circumstances the intensity of the radiant energy at the aperture dis- 
torts the film and causes the picture to go "in and out of focus." This 
has emphasized the need for fundamental information on the intensity 
and spectral constitution of the radiant energy at the film aperture. 

The following discussion deals with this subject and with some 
methods for removing radiant energy which does not contribute to 

* Presented Oct. 16, 1944, at the Technical Conference in New York. 
** National Carbon Company, Inc., Fostoria, Ohio. 


Radiant Energy Intensity Method of Measurement. The 
measurement of the quality and quantity of radiant energy at the 
center of the film aperture was made with the system indicated in 
Fig. 1. Mirror systems as well as the condenser system sketched 
were studied. Prior to making the heat measurements with the 
arrangement shown, the regular optical system including aperture 
plate and projection lens was set up. The entire system was then 
adjusted to give the maximum light intensity at the center of the 
screen. This resulted in approximately 65 per cent as much light 
at the sides as at the center of the screen. After this adjustment 
was made, the film aperture was replaced by a plate having a pin- 
hole slightly more than 2 mm in diameter and so positioned as to 
correspond to the center of the film aperture. The pinhole aper- 






FIG. 1. Diagram of system for measuring radiant energy at the center of 

film aperture. 

ture plate was placed directly in the beam with no shutters, screens, 
or filters between it and the light source. The plate containing the 
pinhole was water-cooled to prevent distortion or damage. The 
thermopile was placed at various positions in the beam, at a suit- 
able distance from the pinhole, as shown on Fig. 1. The thermo- 
pile output was measured by a sensitive galvanometer. To ob- 
tain measurements in absolute units the thermopile-galvanometer 
combination was calibrated using a Bureau of Standards carbon 
filament radiation standard. 

The thermopile with its fluorite window is sensitive to radiation of 
wavelengths from 1700 to 120,000 A. However, absorption by the 
glass of the condensers or mirrors in the optical system limits the 
beam radiation reaching the aperture to wavelengths of 3400 to 42,000 
A. Through the use of quartz water cell and red glass filters, this 
total radiation may be broken down into the following parts : 4 

104 ZAVESKY, NULL, AND LOZIER Vol 45, No. 2 

(1) 3400 to 6300 A 

(2) 6300 to 11,250 A 

(3) 11,250 to 42,000 A 

The thermopile readings obtained in different positions over the 
beam were integrated to obtain both the total energy passing through 
the pinhole and the energy in the 3 wave bands given above. From 
these values and the area of the pinhole the reported values of watts 
per square millimeter were calculated. 

Radiant Energy Intensity Results. A variety of popular car- 
bon arc motion picture projection systems was studied. These 
are listed in Table 1 and ranged from a low-intensity arc at 32 
amp to a super high-intensity arc at 170 amp. The energy meas- 
urements at the center of the aperture were made as just described. 
The results of these determinations are summarized in Table 1 . 

The sources and optical systems listed in Table 1 represent typical 
units used in motion picture projection and include low-intensity, 
copper-coated, nonrotated high-intensity and plain rotated high-in- 
tensity carbon arcs. The data on the total radiant energy in- 
tensity at the center of the film aperture show a range of 3 to 1, from 
1.05 w per sq mm for the 13.6-mm source at 170 amp to 0.35 w per sq 
mm for the 12-mm source at 32 amp. This signifies that there is po- 
tentially 3 times as much heating effect on the center of the film at 
the aperture for the system with 13.6-mm carbons at 170 amp than 
for the system with 12-mm carbons at 32 amp. The measurements 
in Table 1 also show the portion of the total energy in the regions (1) 
below 6300 A (approx. 3400 to 6300 A), (2) 6300 to 11,250 A, and 
(3) 11,250 to 42,000 A. Considering the system with 13.6-mm car- 
bons at 170 amp, for instance, 41 per cent of the total radiant energy 
is of wavelengths shorter than 6300 A and 59 per cent is of wave- 
lengths longer than this value. 

All of the radiant energy absorbed by the film contributes to heat- 
ing, but only the visible energy transmitted by the film is effective for. 
seeing. Since the filter combinations conveniently available do not 
precisely isolate the visible region 4000-7000 A, the per cent of the 
total energy of wavelengths within these limits was calculated from 
previous spectral energy distribution data 2 and from the measure- 
ments of energy in the 3400-6300- A band shown in Table 1. The ra- 
diant energy incident at the center of the film aperture for the projec- 
tion systems listed in Table 1 is composed of approximately 50 per cent 
visible energy for the high-intensity sources and 25 per cent visible 

Aug., 1945 



:> 9 -T a e c 




(M O 

CO "3 

00 00 i* 
CO t t- 


o _ft ^a .S ^ft o 

I IIU : 

8? ^ : : " . : ffi 

^ a a a .s e d 
e a s 2 'ft 

'o x 







i < 








ojector shu 




















106 ZAVESKY, NULL, AND LOZIER Vol 45, No. 2 

energy for the low-intensity 12-mm source, with the remaining energy 
almost entirely within the infrared region. Therefore it can be seen 
that high-intensity carbon arc sources give approximately twice as 
much light per unit of heat as low-intensity arcs. 

Filters for Removal of Infrared. Since approximately 50 per cent 
of the radiant energy incident at the center of the film aperture with 
high-intensity carbon arc motion picture projection systems is 
nonvisible, it can be seen that the removal of this nonvisible energy 
before it reaches the aperture would reduce the potential heating 
effect with no loss in light. Practical means of approaching this 
ideal condition through the use of special filters forms the subject 
of this portion of the paper. Theoretically the removal of all radi- 
ant energy of wavelengths shorter than 4000 A and longer than 
7000 A would give the best results in reducing heating effects with- 
out decreasing light. However, no filters practically available 
have cutoffs so precise, although they do furnish methods whereby 
a significant portion of the nonvisible energy may be removed. 

Some liquid and some glass filters have been studied regarding their 
light and heat transmission properties, and regarding the amounts of 
energy they remove in the particular spectral regions listed in Table 1. 
These data are given in Table 2. A 2-in. thick layer of distilled water 
contained in a Pyrex cell removes about 15 per cent of the light and 
40 per cent of the total energy from the beam of a high-intensity car- 
bon arc motion picture projection system. A 3-mm thick sample of 
395 Extra Light Shade Aklo or 2043X Phosphate glass 5 removes 20 to 
25 per cent of the light and 45 to 55 per cent of the total energy from 
such beams. 

An important criterion of filter performance is its ability to remove 
as much nonvisible energy and to transmit as much visible energy as 
possible. The ratio of light and total energy transmission gives a rela- 
tive measure of the amount of light per unit total radiant energy 
passed through the various filters. On the basis of 100 for the unfil- 
tered beam, the amounts of light per unit total energy vary, as shown 
in Table 2, from 140 for the distilled water and the Aklo filters to 160 
for the phosphate glass. This means that the quantity of light could 
be increased by 40 to 60 per cent without any increase in total energy, 
with sources having the quality of radiation of the 8-mm Suprex and 
13.6-mm super high-intensity carbons listed in Table 2. 

The water cell and the glass filters absorb a much greater propor- 
tion of invisible energy of wavelengths longer than 6300 A than they 

Aug., 1945 



per sq 


i-H O O O 

do do 

^^ C^ t^" CO 

do do 


d d 
ao oo" 


d d 


o o o o 

TP T^ 

-tf CD 


00 00 



d d 


d d 

1-1 O O o 

lags o 
<*** *- 



o o 

Tf CD 




^ 88 



fe S 
%* 8- 

L^ Q^ O 


Q aj co ^ 




% 3 

^s < 


fc <N 

X Phospha 

N. 00 


do of shorter wavelengths. FOE example, the phosphate glass removes 
about four-fifths of the energy of wavelengths longer than 6300 A and 
only one-sixth of the portion of wavelengths shorter than this value. 
This results, for instance, in the case of a phosphate glass filter, in 
filtered beams having 75 per cent or more of the total energy of wave- 
lengths shorter than 6300 A compared with 40 to 45 per cent of the 
total energy in this wavelength region for the unfiltered beam. The 
difference in quality and quantity of radiant energy between the un- 
filtered and filtered beams is given in Table 2 for each of the filters dis- 
cussed. Of the filters listed, the water does not change the color tem- 
perature of the projected light, but the Aklo and Phosphate glass in- 
crease the color temperature slightly. 

The use of filters which remove most of the radiant energy of wave- 
lengths longer than 7000 A before the projected beam is concentrated 
at the film aperture can facilitate the utilization of improved carbon 
arc projection systems capable of producing more light on the screen. 


1 BOWDITCH, F. T., AND DOWNES, A. C.: "Spectral Distributions and Color 
Temperatures of the Radiant Energy from Carbon Arcs Used in the Motion 
Picture Industry," J. Soc. Mot. Pict. Eng., XXX, 4 (Apr., 1938), p. 400. 

2 NULL, M. R., LOZIER, W. W., AND JOY, D. B.: "The Color of L-ight on the 
Projection Screen," /. Soc. Mot. Pict. Eng., XXXVIII, 3 (Mar., 1942), p. 219. 

3 CARVER, E. K., TALBOT, R. H., AND LOOMIS, H. A.: "Effect of High-In- 
tensity Arcs Upon 35-mm Film Projection," /. Soc. Mot. Pict. Eng., XLI, 1 (July, 
1943), p. 69. 

4 COBLENTZ, W. W., DORCAS, M. J., AND HUGHES, C. W.: "Radiometric 
Measurements on the Carbon Arc and Other Light Sources Used in Phototherapy," 
Scient. Papers of the Bur. of Stand., 21, 539 (Nov., 1926), p. 535. 

5 ,9,95 Extra Light Shade Aklo supplied by Corning Glass Company ; 2043X 
Phosphate by Pittsburgh Plate Glass Co. 


Summary. A machine has been designed and built for edge-numbering 16 -mm 
original and work prints. The numbers are printed with white ink so that they may 
be read easily on the black edges of reversal or color film. It is 'estimated that by 
using the machine the matching of originals with work prints can be done in about 
one-fourth the time formerly required. The service is available to anyone making 
pictures by the direct 16-mm method. 

Producers of direct 16-mm pictures have long recognized the value 
of a machine which could be used for edge-numbering original photog- 
raphy, work prints, sound tracks, or prints of sound tracks. There 
have been numerous proposals made that 16-mm film be edge-num- 
bered with a latent image, but the film manufacturers have never 
gotten around to providing this except on special order. 

Increased production of direct 16-mm photography and sound 
made the editing problem so acute that The Calvin Company de- 
cided to build an edge-numbering machine which could be used for 
edge-numbering original photography, work prints, and sound tracks 
after they had been developed. This system has been in use only a 
few months, but we have found that it has cut the time of matching 
originals to work prints by approximately 75 per cent, and that such 
work can now be done by relatively untrained persons. 

The machine was made with a standard numbering head such as is 
used for edge-numbering 35-mm film. It is for this reason that the 
numbers which are printed along the edge of the film now are back- 
ward. We have on order a special numbering head with the numbers 
reversed so that the number will appear right-side up when printed 
along the edge of a black-and-white reversal or color film. However, 
this is not a serious problem as the person doing the editing soon 
learns to read the numbers almost as quickly as though they were 
right-side up, much the same as a printer reads type. 

* Presented Oct. 17, 1944, at the Technical Conference in New York. 

* The Calvin Company, Kansas City, Mo. 


110 L. THOMPSON Vol 45, No. 2 

The numbers which are printed along the edge of the film are 
printed in white ink so that they can be easily read. When the 
standard purple or black numbering ink is used it has been found very 
difficult to read these numbers on black-and-white reversal film or on 
color film. A special ink was developed which prints legibly on the 
film and dries quickly so that there is no tendency to smear. Further- 
more, if the numbers must be removed from the film they can be re- 
moved easily. However, they do not rub off during the normal edit- 
ing process. 

FIG. 1 . A 16-mm edge-numbering machine for numbering developed 


The numbers are accurately printed between the sprocket holes so 
that every number is legible and not blotted out by a perforation in 
the film. Furthermore, the machine has an accurate adjustment to 
compensate for shrinkage if the work print should be shorter or longer 
than the original. In other words, edge numbers can be printed by 
this method as accurately as though they were latent numbers on the 
original. If the latent image method is used in edge-numbering 
16-mm film, it happens quite frequently that some numbers will fall 
in a sprocket hole and this sometimes makes them hard or impossible 
to follow. 

Aug., 1945 



It is assumed that producers of direct 16-mm photography will 
want to use work prints which have been made by the reversal method 
so as to give a positive picture image. Therefore the white ink is also 

FIG. 2. Original 16-mm photog- 
raphy with edge numbering as it will 
appear when marked with the number- 
ing head described in the text. 

used for edge-numbering work prints. Sound tracks usually have 
clear edges, but the white ink can also be read on them. 

There are some advantages to edge-numbering 16-mm film after it 
has been shot instead of using the latent image. In the first place, it is 
rather difficult to get film which is edge-numbered with a latent image 
and there is always a chance that a few rolls will be shot in a produc- 



tion which have not been edge-numbered. This, of course, is almost 
worse than no edge-numbering at all. In editing a 16-mm "show" it 
is customary to work in lengths of one reel or 360 to 390 ft. Therefore 

when the original photography is spliced 
together to make work prints it is spliced 
in lengths not longer than 390 ft. The 
reels are then printed. If we have a 
1200-ft show, then we have 3 reels of 
original photography and 3 reels of work 
print before any cutting has been done. 
The work prints are synchronized with 
the original photography and the first 
reel of photography is numbered from 
one to 390 ft. The work print is then 
numbered with exactly the same numbers 
as were used on the original photography. 
The second reel of original photography 
is then numbered from 400 to 790 ft, and 
the work print numbered the same way. 
The third reel is then numbered from 800 
to 1190 ft, and the work print numbered 
in the same manner. The work print is 
then cut and after all the work has been 
done with the work print it is ready for 
the cutter to match the original photog- 
raphy with the work print. 

In editing the picture, let us suppose 
that one of the scenes which was in the 
third reel was put in the first reel. By 
looking at the number on the work print 
the editor knows immediately that the 
scene was taken from the third reel of 
original photography because it carries 
the number 1052ft. With this informa- 
tion it is no trouble at all to locate the 
proper scene and put it in its proper place in the original pho- 

FIG. 3. A 16-mm sound 
track as it may appear if the 
recording is done on "A 
Winding" stock. 



Summary. In the early days of Edison's work with motion pictures it seems 
that he, too, would indulge in that kind of daydreaming which begins "Wouldn't it 
be wonderful if ." There is ample evidence that the sentence ended with the words, 
" we might have both color and sound in educational motion pictures." Edison's 
daydreaming is a reality today, if we choose to use the materials and processes already 

Kodachrome can be considered a successful process. Although its photographic 
speed is slower than black-and-white films, it is almost as convenient to use in the 
ordinary 16-mm camera. Like all color processes, it has its limitations which, if 
ignored, may lead to unnecessary disappointment. For most uses, these limitations 
can be avoided. 

It must be recognized that there is no "perfect" color process. The usual require- 
ments for a satisfactory color process include: 

(1) A suitable gray scale, * ' 

(2} Comparable color scales for the components, 

(3) Accurate reproduction of color, 

(4) Good differentiation of color. 

Unfortunately each of these requirements conflicts with at least one of the others. 
Ordinarily (1) and (4) are favored over (3); the result of this compromise is satis- 
factory for most purposes. 

In medical work where accurate reproduction of color is often desired for diagnostic 
and similar purposes, some of the very common biological stains are not reproduced 
accurately -in integral tripack color films. In such cases and in other specialized 
cases where the absorption spectra of the photographed material are "unfortunately" 
located, color accuracy must knowingly and intentionally be sacrificed for color 

Some data on films and filters not previously published are included. Much of this 
has been in use commercially for several years and has been helpful in solving in a 
practical way the everyday problems of color duplicating that arise in commercial 
laboratory work. Mention is made of some of the fundamental color standardization 
accomplished and its relationship to commercial duplicating in the independent 

* Presented Oct. 17, 1944, at the Technical Conference in New York. 
** New York. 


114 W. H. OFFENHAUSER, JR. Vol 45, No. 2 

Years ago when color in motion pictures was just a matter of scien- 
tific speculation, Thomas A. Edison expressed the thought that the 
combination of color and sound in motion pictures would represent 
the highest pinnacle in motion picture technological achievement. 
By Edison's criterion, it would appear that we have already arrived 
at that millennium. 

As in the case of prior arts, the daydreams of the pioneers became 
the realities of a later day. The embodiment of a daydream usually 
brings forth new problems that must be solved in turn. We may view 
this as the secondary stage in the development of an art. And so it is 
with integral tripack color films; they are entering the stage of 
application and intensive secondary technological improvement. 

The idea of multilayer films for color purposes traces back to the 
earlier stages of the motion picture long before sound became a com- 
mercial feature. The dye-coupler concept likewise harks back to an 
early time ; its potentialities were appreciated to a remarkable degree 
as early as 1907. The names of Homolka, 1 Lewy, 2 and Fischer 3 
will be remembered as early pioneers with vision. 

It has taken a long time for the processes described by these in- 
vestigators prior to 1914 to become commercial realities. Just as the 
photoelectric cell had to wait for the coming of electron tubes and 
amplifiers before sound films could be commercialized, the integral 
tripack color films had to wait for suitable dyes and other chemicals. 
As in other fields of invention, each new discovery added to the total 
store of knowledge and, in turn, each new discovery was found to have 
its limitations. It appears that there is no solution that will meet all 
the requirements of an "ideal" color process; there is no "perfect" 
film or "perfect" process. 

Kodachrome Processing and Duplicating: Some History. 
Additive processes such as the old Kodacolor process gave way com- 
mercially to subtractive processes. No filters or other gadgets are 
needed for either camera or projector when integral tripack sub- 
tractive color films are used; this was considered a "must" for color 
film for the amateur. Kodachrome introduced in 1935 is the 
most widely used 16-mm motion picture color film. It was made 
simple for the user despite the fact that it was complex in manufacture 
and in processing. When it was first introduced the developing proc- 
ess included approximately 30 stages, each of which required precise 
control. The process worked well; amateurs and others bought film, 
exposed it in their cameras, and projected the results in color. 


To an operator of a commercial film laboratory, the requirement 
that a particular piece of film must go through some 30-odd control 
stages automatically brings the reaction "I'm glad that it is the 
manufacturer's problem and not mine." In black-and-white com- 
mercial laboratory work we still seem to have considerable difficulty 
today with but a fraction of that number of control stages. 

It was logical to expect the Eastman Kodak Company to simplify 
such a complex developing process in order to cut operation costs and 
to reduce the risks of damage during processing. Such simplifica- 
tions had to be introduced while commercial film was being processed 
daily; if the customer were to be aware of the difference at all, he 
should observe it as an improvement. Obviously something in the 
process had to be "tied down." It is fair to say that the average user 
was not aware that changes in the process were constantly being made; 
yet there were unmistakable signs that Kodachrome was improving. 
There was close control of emulsion manufacture and of color de- 
veloping as well as coordination between them. The price at which 
the raw film was sold included the developing cost; there was little 
opportunity on the part of the manufacturer's laboratory to "pass 
the buck." (No commercial laboratory attempted to color-develop 
Kodachrome; no such laboratory regardless of its personnel and 
equipment could ever hope to operate at a profit under such strict 
control requirements especially since the developing cost was already 
included in the price of the raw film and when the process itself was in 
a "fluid" state.) 

Soon after Kodachrome first made its appearance, it was logical to 
expect that attempts would be made to duplicate it. In the early 
stages, duplication was little more than placing Type A raw film in a 
contact printer with the original picture and then snapping the switch. 
The printed film was hopefully shipped to Rochester; occasionally a 
developed roll with images would return. Oftentimes there was 
an apologetic letter together with a new roll of raw film. 

When commercially usable films started to return with some de- 
gree of frequency from Rochester, a commercial duplicating busi- 
ness was born. Even at that early stage the advantages of color 
and sound in 16-mm were appreciated; a survey of certain indus- 
trial film users showed that more than 90 per cent of those canvassed 
wanted color in their 16-mm sound films and only Kodachrome was 
able to give it to them. The rapid growth of Kodachrome duplica- 
tion, therefore, was not entirely unexpected. 

116 W. H. OFFENHAUSER, JR. Vol 45, No. 2 

The Available Types of Kodachrome. Kodachrome for the 16- 
mm camera is sold in 2 different color balances : Regular, for day- 
light, code EK 5263 balanced for 6100 K source temperature; Type 
A , for Mazda, code EK 5264 balanced for 3450 K source tempera- 
ture. The price at which the film is sold includes color developing. 

Kodachrome for duplicating is sold in a single color balance: 
Duplicating, for 2900 K source temperature, code EK 5265. As in 
the case of Kodachrome for the camera, the price includes color de- 
veloping. Sixteen -millimeter Kodachrome is designed to be used 
with a specified arrangement of filters for "balancing" purposes and 
with the lamp operated at the specified temperature of 2900 K. 
Further data upon the process will be given later in this paper. 

The duplication of Kodachrome has grown in volume to the point 
where each of several laboratories not connected with the film manu- 
facturer is printing several million feet per year. The total handled 
by all users is a truly large item when we remember that the average 
cost is approximately 8 cents per ft compared with approximately 
1 Y 2 cents per ft for black-and-white. 

Competitive Positions of Present-Day 16-Mm Color Methods. 
It would seem worth while at this point to give some thought to the 
future as signs point to a still further increase in volume of 16-mm 
color film. For most purposes, the present Kodachrome product is 
quite satisfactory. There are several competitive factors, however, 
that are the imponderables of the future. 

First, there is the marketing of Ansco Color Film. At first glance, 
Ansco Color Film would seem to be in the fortunate position of shar- 
ing the future 16-mm market with Kodachrome. Undoubtedly the 
color printing techniques will be similar. With good sensitometric 
control, color development should not encounter any very serious 
obstacles although most prospective users will have to begin to learn 
what quality control and process control really mean if they hope to 
be commercially successful. This is a very real and serious problem ; 
even the most optimistic of us would hardly dare to say that good 
control has been achieved when we screen prints of training films for 
the Armed Forces. 

The second imponderable is sound. The sound quality available 
with present-day Kodachrome does not compare with the quality 
obtainable under properly controlled conditions with high resolving 
power films such as the blue-dyed EK 5372, or the yellow-dyed EK 
5365. (The rated resolving power of both is 150 lines per mm.) 


Bruno 4 does not hold out much hope for conventional methods o 
improvement when he states that both Ansco Color Film and Koda- 
chrome exhibit resolving power of the order of 40 lines per mm. It 
would seem that the difference between Bruno's measurements and 
the Kodak ratings can be probably attributed to a difference in 
measuring technique. It would be desirable to evolve some empirical 
method for evaluating resolving power. This would avoid apparent 
discrepancies; to paraphrase Mark Twain, "An argument arises when 
two people use the same words to describe different things." Pos- 
sibly 2 figures might be established one representing the perform- 
ance of the picture portion of the film (this could establish agreement 
upon a single visual method), and the other representing the perform- 
ance of the sound portion of the film. It would seem that the sound 
art has already reached the point where some such evaluation 
method could be used for expressing the performance of a sound 
record when scanned by the scanning beam of a projector. Such 
evaluations would be useful in comparing the performance of black- 
and-white with color film; a standard projector will project either. 

In a recent issue of the JOURNAL, Gorisch and Gorlich 5 discuss some 
of the criteria for a satisfactory sound track on multilayer films and 
report upon some of their tests. Their conclusion that the usual 
caesium surface photoelectric cell is well suited to the reproduction of 
silver-emulsion films, but should be modified with antimony if it is to 
reproduce dyed films satisfactorily, is significant. No doubt much 
undisclosed progress has already been made in making the cell to fit 
the film and the film to fit the cell. While the general trends that the 
characteristics should take seem indicated, the problem is really a 
knotty one and will require considerable further thought before it is 
satisfactorily solved in terms of the high-resolving power of 90 lines 
per mm required of projection lenses. It may well turn out that the 
major part of the 16-mm color film market will be initially awarded 
to the film manufacturer whose product delivers outstanding sound 
quality as the color problems for the picture seem less difficult of 

The third imponderable is picture detail. The loss in detail in a 
very well-made Kodachrome or Ansco Color duplicate is excessive 
when we think in terms of lenses with 90 lines per mm resolving power 
used on projectors with 50 amp on the arc throwing light upon 15-ft 
screens. Low resolving power is inherent in multilayer films as it is 
necessary that relatively fast individual emulsions be used for even a 

118 W. H. OFFENHAUSER, JR. Vol 45, No. 2 

slow-speed final product. Bruno 4 reports an immediately available 
increase from 40 lines per mm to 55 lines per mm if certain sacrifices 
are made in color rendition. In the making of duplicates, Bruno 
further reports that the number of filters used in printing will have 
to be kept to a minimum as there is evidence of considerable loss of 
resolution (about 10 per cent) for every filter introduced in an imag- 
ing optical system.* 

The fourth and final imponderable is the competitive position of 
imbibition printing. (Technicolor is an example.) Where the num- 
ber of prints in an order runs to 500 or more, imbibition printing 
should show some very generous profits at the present duplicate bulk 
price of 8 cents per ft. An additional factor in its favor is that the 
resolving power problem does not loom so large when compared with 
integral tripacks. Laboratories that duplicate Kodachrome have 
experienced 500-print orders, and it is likely that they will fight hard 
to boost quality as imbibition color printing enters the field as a 
strong and able competitor. 

The foregoing and other marketing factors seem to point to the 
conclusion that 16-mm color-sound printing will probably become 
quality-competitive and price-competitive at the same time. Be- 
cause of the importance of production methods and their influence on 
production costs, some notes upon the present Kodachrome dupli- 
cating techniques are indicated. 

Picture Duplication. The general instructions for making 16-mm 
Kodachrome duplicates are found in a booklet "Instructions for 
Making 16-Mm Kodachrome Duplicates on Kodachrome Duplicating 
Film Code 5265," Issue No. 8, March 30, 1944. This should be in 
the hands of all those engaged in the business of making Kodachrome 
duplicates. It is issued by the Motion Picture Film Department of 
the Eastman Kodak Co. 

The basic method is simple. The lamp in the printer is first set 
and maintained at the color temperature of 2900 K. The intensity 

* The gelatin compensating filters of the CC series supplied by the Eastman 
Kodak Company for Kodachrome duplicating are considered "optically inert" 
and do not introduce losses as serious as the more permanent glass filters. They 
are, however, somewhat unstable when subjected to heat and must be checked 
periodically when they are used in commercial duplicating. The loss in resolution 
caused by a particular filter in an imaging system varies quite widely when used 
in different parts of an optical system. Barring heating and similar deteriorat- 
ing effects, correction filters are best used over the light source as it is here that 
they introduce minimum loss in resolution. 


is then reduced to the appropriate value by means of neutral density 
niters or equivalent after the specified printing filters are mounted in 
place in the light beam.* A test original is then printed; production 
prints are made after the minor corrections indicated by the test are 
made. All tests should be made upon the same emulsion lot of film 
as that which will be used for production printing. Slightly different 
''balances" will be required for different emulsion lots. In other 
respects, the printing process is comparable to black-and-white 

Some Limitation of the Duplicating Process. When certain films 
are duplicated in the specified manner, the result on the screen may 
leave something to be desired. There are several possible sources : 

(1) An accurate copy of the original may not be desired; the color balance of the 
original may be quite different from that desired in the duplicate. It often happens 
that colors "ideal" for Kodachrome are not possible or convenient in the original 
photographing; photomicrography, where bacteria dyes and stains are involved, 
is a typical example. 

(2) The exposure of the original may be quite different from that desired in the 
duplicate. One example would be the exposure required for a short sky shot 
interposed between 2 shots of dark woodland; such a sequence might be found in 
a training film dealing with the landing of airborne troops. 

(3} The duplicating process itself has limitations. Some practical duplicating 
problems cannot be solved satisfactorily by merely following the printed Kodak 
instructions; they require an understanding of the pertinent limitations of the 

The Theoretical Elements of the Color Process. As a starting 
point, we may say that the color spectrum has been arbitrarily (yet 
with good reason) divided into 3 major parts. For the purpose of 
duplicating, a color compensating filter may be considered to provide 
attenuation in only one part of the spectrum. To provide different 
amounts of attenuation, each type of filter is made in different densi- 
ties. For convenience, the color densities can be the same in all 3 
parts of the spectrum. We would find, therefore, a series of minus- 
blue filters, a series of minus-green filters, and a series of minus-red 
filters. Eastman Kodak has chosen 3 color densities 0.06, 0.12, and 

* The specified filters are: 

2 3.2-mm Aklo (supplied by Kodak) 
1 Wratten 2 A 
1 Kodak CC45 
1 Kodak CC44 
1 Kodak CC34 

120 W. H. OFFENHAUSER, JR. Vol 45, No. 2 

0.22 as the rated color densities for each spectrum part. Since each 
filter attenuates in only one-third of the spectrum, the equivalent 
neutral density for a particular color density may be considered one- 
third of its respective color density. Thus the equivalent neutral 
densities for the color densities chosen are 0.02, 0.04, and 0.07, respec- 
tively. Roughly speaking, the minus-blue (yellow) range extends 
from 400 to 500 m^u ; the minus-green (magenta) range from 500 to 600 
m/z; and the minus-red (cyan) range from 600 to 700 mju. Fig. 1 
shows the filters and their ratings in tabular form. 

Friedman 6 presents an interesting yet simple discussion of the first- 
order relationships of what may be termed the 3 components of a 3- 
color process. Theoretically a set of compensating filters such as 
those described should be adequate for the purpose according to the 
criteria for color reproduction set forth. In practice, the Kodak set 

Attenuation Ranges 


400 to 500 mM 

500 to 600 m/j, 

600 to 700 m/i 









FIG. 1. Ratings of Eastman color compensating filters for Kodachrome dupli- 

of filters is adequate in most cases, but there is a small number, par- 
ticularly those cases in which large-order corrections are to be intro- 
duced, in which the simplified Kodak procedure leaves something to 
be desired. A study of the process should explain the discrepancies. 

Color Standardization. As a starting point, we have a standard 
called "The Specification and Description of Color." This was 
issued as American War Standard Z44-1942 by the American Stand- 
ards Association. Section 2.1 reads "The spectrophotometer shall 
be recognized as the basic instrument in the fundamental standardi- 
zation of color." A convenient instrument that is commercially 
available is the General Electric Recording Spectrophotometer. 
This instrument is widely used; it is available in commercial testing 
laboratories such as Electrical Testing Laboratories in New York. 
Curves shown in this paper were taken on this instrument. 

Gage 7 reviews color measurement and sets forth some of the terms 
regularly used to describe the attributes of color. There is a generous 


list of references at the end of the paper for those who are interested 
in the many scientific aspects of color. 

No discussion of color standardization, however brief, can be con- 
sidered complete without mention of the ICI Standard Observer and 
Coordinate System for Colorimetry. Color specifications prepared 
in accordance with this internationally accepted method can be com- 
puted from spectropho tome trie data. Unless otherwise specified, 
standard ICI illuminant C (representative of average daylight) is 
assumed. Results of the computations are expressed in a table of 3 
values for each wavelength in the spectrum. These data are then 
plotted as a curve for convenience. Owing to the computations re- 
quired, this method of describing color is limited in its use ; for most 
purposes curves taken by the recording spectrophotometer are pre- 

The "Munsell Book of Color" must also be mentioned as a catalog 
or atlas of color that is in wide use. This book is made up of a large 
number of color samples. The characteristics of manv of these have 
been measured on the recording spectrophotometer and also trans- 
lated into ICI terms. The colors in this book were used as a standard 
prior to 1931 at which time the ICI formulated the present system. 
The Munsell Book is still of real value as a reference not only because 
of the great variety of colors included, but also because of the uni- 
formity and* the permanence of the materials used for them. 

The Textile Color Card Association of America has taken the lead in 
recommending the use of standard names for colors used in the textile 
industry. The Association has not only selected suitable names for 
the colors but has also been analyzing each color spectrophoto- 
metrically. It is not uncommon for specifications to describe the 
colors of radio hookup wire insulation and of color coding of the re- 
sistance values of resistors used in radio equipment in terms of the 
Textile Color Card Association colors. 

Mention must also be made of the influence of the graphic arts in 
color standardization, particularly the "Offset Color Guide" pub- 
lished by the International Printing Ink Division of the Interchemical 
Corporation. This guide contains over 100 separate test frames (of 
the same subject) and not only illustrates a large variety of color 
shades arranged according to dominant color, but also gives the 
specification for each illustration in terms of both the spectrophoto- 
metric and the Munsell color factors in accordance with ASA Standard 
Z44. This booklet represents a convenient and reliable source of 



Vol 45, No. 2 

color illustrations for color photographing and like tests. Mention 
must also be made at this point of the "Three Monographs on Color" 
published by the same company. These volumes are delightful in 
addition to being scientifically correct. They should be of interest to 
both scientist and layman alike interested in color in whatever aspect. 







5OO 600 



FIG. 2. Transmittance- wavelength characteristics of 
Eastman minus-blue (yellow) compensating filters. 

Color Filter Criteria. One of the first items to be investigated is 
the transmission characteristics of the filters used. To the elec- 
trical engineer with communications experience, the mention of the 
word "filters" brings 3 questions to mind: 

(1) What are the actual transmission characteristics in the pass band? 

(2) What are the attenuation rates at cutoff and at crossover? 


(3) What are the transmission "discontinuities" and irregularities? 

The concept of Q is so firmly established in the electrical engineer's 
mind that he would prefer to think of optical filters in the same man- 
ner. Although we have not yet learned how to design optical filters 


500 600 



FIG. 3. Transmittance- wavelength characteristics of 
Eastman minus-green (magenta) compensating filters. 

with knobs on them that will permit ready adjustment of their resist- 
ance, inductance, and capacitance equivalents, the electrical engineer 
is not stopped from thinking of their performance in such equivalent 

Sources of Filter Data. Optical filter data are obtained from the 
catalogs of manufacturers and from such references as the Hodgman- 
Holmes "Handbook of Chemistry and Physics," familiar to most 

124 W. H OFFENHAUSER, JR. Vol 45, No. 2 

students of chemistry. Some of the more important catalogs include : 

(1) "Glass Color Filters," Corning Glass Works, Corning, N. Y. (Form 

(2) "Jena Colored Optical Filter Glasses," Fish-Schurman Corp., New York 

(3) "Neue Lichtfilterglaser," Fish-Schurman Corp., New York (5990g). 

(4) "Wratten Light Filters," Eastman Kodak Company, Rochester, N. Y. 
(16th Ed.). 

Data for Filters Used in Kodachrome Duplication. The per cent- 
transmission wavelength characteristic of the 2 A Wratten filter is 
given in "Wratten Light Filters" and the characteristics of Aklo 
filters are given in "Glass Color Filters," both mentioned above.* 

Data for the Kodak Color Compensating Filters have not been pub- 
lished previously ; Fig. 2 shows the characteristics for the minus-blue 
series (CC23, CC24, and CC25), Fig. 3 shows the characteristics of the 
minus-green series (CC33, CC34, and CC35), and Fig. 4 shows the 
characteristics for the minus-red series (CC43, CC44, and CC45). 

If these various sets of color compensating filters are compared, 
significant differences are apparent. Although each set represents a 
"family," only the minus-blue series seems to fit the "ideal" criterion 
that attenuation shall take place only in the band for which the filter 
is rated. Even this series does not have a "square wave" transmis- 
sion characteristic. It would be unreasonable to expect such a 
characteristic from the coloring materials available for filter making. 
Since the differences, though small, are significant, it is well to sum- 
marize them. 

Minus-Blue Series (CC23, CC24, CC25) 

(1} One principal absorption wavelength 430 m^u. 

(2) Negligible attenuation in either the minus-green or the minus-red bands. 

Minus-Green Series (CC33, CC34, CCS 5} 

(1} Two principal absorption wavelengths 530 and 565 mju. 
(2} Slightly greater attenuation at 565 m/z than at 530 mju. 

(3) Slight attenuation in the minus-blue range (approximately 10 per cent at 
430 m M for CC35). 

(4) Negligible attenuation in the minus-red range. 

* The 2 pieces of Aklo glass supplied by Eastman Kodak are 3.2 mm. thick. 
The data given in the Corning catalog for Aklo numbers 3966 (extra light shade), 
3965 (light shade), 3962 (medium shade), and 3961 (dark shade) are for a thickness 
of 2 mm. 

Aug., 1945 



Minus-Red Series (CL43, CC44, CC45) 

(1) Broad band nonsymmetrical transmission centered about 500 m/j.. 
(2} Slight attenuation in the minus-blue band (approximately 15 per cent at 

430 m M for CC45). 

(3} Greater attenuation at 565 m/i in the minus-green band than at 530 m/z 

(almost 15 per cent). 









FIG. 4. Transmittance- wavelength characteristics of 
Eastman minus-red (cyan) compensating filters. 

(4) Appreciably greater attenuation at the longer wavelength end of the minus- 
red band than at the shorter wavelength end (approximately one-third greater for 
CC45 at 700 than at 600 m^). 

Should it be necessary to use other filters for correction purposes it 
will be found that it is not convenient to compare quickly the pub- 
lished characteristics of filters made by different manufacturers owing 
to the differences in the manner in which such characteristics are pre- 



Vol 45, No. 2 

sented. It is expected that in the near future all manufacturers will 
use standard scales for published data and standard methods of 
measurement so that optical filters for photographic purposes may be 
readily compared regardless of their origin. Progress has already 
been made in this direction by the War Committee on Photography 
and Cinematography of the American Standards Association. 

Data for Kodachrome Duplicating Film. The film itself can be 
checked for deviations from the "ideal." Fig. 5 shows a per cent 
transmittance versus wavelength characteristic of unexposed de- 
veloped Kodachrome; for comparison purposes both Type A and 


o w ^ o> o> c 




)0 500 600 70 


FIG. 5. Transmittance- wavelength characteristics of 
unexposed and developed Kodachrome: No. 5 EK 
5265 (Duplicating); No. 6 EK 5264 (Type A). 

duplicating Kodachrome are shown. From these curves it is reason- 
able to conclude that Kodachrome when unexposed provides a satis- 
factory neutral gray. 

The film can next be checked under simulated operating conditions 
Fig. 6 shows these characteristics when a fine-grain black-and-white 
film, which is uniformly exposed (EK 5365 exposed and developed to 
a density of 0.5), is printed on a step-contact printer through the 
basic set of filters recommended by Eastman Kodak for duplication. 
These curves show that at the exposure used the neutral gray is rea- 
sonably well maintained. In addition, they indicate that although 
EK 5264 Type A Kodachrome requires less exposure than EK 5265 
duplicating Kodachrome, the duplicating film has the smoother curve 


and should be used wherever possible. Among other reasons for 
preferring it, is the not unimportant factor that the price is appreci- 
ably lower. 

Contrast Control. It may well be said that integral tripack 
duplication makes no provision for the purposeful control of con- 
trast. While it is true that in the earlier stages there was little need 
owing to the limited useful contrast range, subsequent improvement 
has brought forth that need. 

The problem is a very difficult one with many facets. It means 
the introduction of still another variable into a process where the 
number of variables is already large compared with commercial black- 
and-white processes. At first glance Ansco Color Film would seem 
to present the possibility in the color developing of the film, but so far 
there has been little encouragement in this direction. In the case of 
Kodachrome the need for constant developing is considered very im- 
portant and it is felt that contrast control is better accomplished in 
some other manner. 

A number of solutions to the contrast control problem has been 
suggested and a certain amount of laboratory experience has been 
obtained with one promising possibility masking film.* So far, the 
complexities of use are such that it has not been considered suitable 
for motion picture use although it is well suited for still pictures. 
No doubt there will be some modification of the basic idea in the 
future that will be satisfactory. It would not seem impossible to 
incorporate this in the form of an additional sensitized layer in the 
duplicating raw stock. Such a layer might have the further ad- 
vantage of being suitable for the sound record thereby avoiding some 
of the seemingly insurmountable problems associated with the multi- 
layer color sound track. 

Production Quality Control of Prints. There is much to recom- 
mend the routine testing of every roll of film printed. It is necessary 

* Masking film when so used is an unexposed sensitized thin film cemented or 
otherwise attached to the developed original Kodachrome picture. Each scene is 
given a predetermined exposure in a printer and then developed in a black-and- 
white developer bath, producing a complementary negative of the picture on the 
masking film as a mask. When the picture is then printed (with the masking film 
still attached), the contrast of the original is effectively reduced. (Should an 
increase in contrast be desired, this might be accomplished by developing the 
masking film as a reversal. It is rare that an increase in contrast is desired; most 
duplicated films suffer from excessive contrast rather than too little contrast.) 



Vol 45, No. 2 





400 500 600 700 


FIG. 6. Transmittance- wavelength characteristics 
for the Kodachrome duplication of a black-and-white film 
with EK 5265 and EK 5264. The conditions of test were : 
(7) Original used: EK 5365 uniformly exposed and de- 
veloped to a density of 0.5. (2) Kodachrome used: 
No. 1 EK 5265 Duplicating Kodachrome; No. 2 EK 
5264, Type A, Kodachrome. (3} Filters used: one 
Wratten 2A, one CC45, one CC44, one CC34, two 3.2-mm 
Aklo. (4) Light source: T-10, 105-v, 500-w lamp at 
approximately 80 v. (5) Printer: DeBrie (contact-step 
printing at approximately 20 ft per min.) 


to establish quality control in printing and to maintain control once 
it has been established. Routine testing is required for every other 
element entering into the process. Commercial color duplication in 
Kodachrome is therefore exacting and demanding, yet interesting 

The starting point is a lamp, the color temperature of which is 
known. The color temperature is not known, however, unless the 
limits to which it has been measured are also known. The same- 
electrical testing laboratories that provide recording spectrophotom- 
eter services can also determine the current at which a particular 
lamp will reach the desired color temperature. If a lot of 50 lamps 
of the same type is tested, the cost for a measurement within a toler- 
ance of == 10 K is but a fraction of the cost of the lamp itself. Should 
seasoning or aging be required, these laboratories can provide such 
service at very low cost. With such reliable services available at 
low rates, no commercial laboratory engaged in the duplication of 
color film, such as Kodachrome, can afford to be without control aids. 
Fig. 7 shows a color-temperature versus voltage versus current rela- 
tionship of a typical 500-w, 105-v T-10 printing lamp. Other valu- 
able lamp data are found in the bulletin "Mazda Lamps." 8 

Exposure is most readily measured directly at the printer aperture 
with the machine stationary ; the better grade illuminometers such as 
the Weston Model 628 (priced approximately $100) are suitable. In- 
direct methods, such as the use of black-and-white film as a control, 
are not practicable in most commercial laboratories as they intro- 
duce additional variables whose variations are usually unknown.* 
In using illuminometers and similar direct -measuring instruments, 
it is necessary to make up suitable jig-adapters so that readings 
of the instrument will be reproduced without significant personal 
error. Provision should also be made for periodically calibrating 
the measuring instruments. It is necessary that the variation in 
slip of belt-driven printers be known. It is still better to use direct- 
gear drive with a 3-phase synchronous motor or equivalent to elimi- 
nate machine speed variation as a possible source of exposure varia- 

* Such indirect methods can be satisfactory if the sensitometric control on black- 
and-white is of a superior order compared with usual commercial control. In 
such cases, however, the direct methods are still more convenient and more 



Vol 45, No, 2 

The regulation of the current supply for the printing lamp is very 
important. It is obviously a sheer waste of time and money to cali- 
brate lamps to =t 10 K when the motor generator or other supply has 
poor regulation, and other loads are indiscriminately "placed on the 













f\ * 








J r\ 















50 60 

70 80 90 100 HO 

FIG. 7. Typical color temperature-volts-amperes characteristics 
of a 500-w, 105-v T-10 lamp. 

line" and "taken off/' A simple check is to connect a recording 
voltmeter to a printer under suspicion, using that printer without 
current changes meanwhile. Often the local power company will be 
glad to lend a suitable meter if one is available. 

From the user's point of view, Kodachrome duplicating film (EK 
5265) is quite consistent. The major variation that occurs is the 
variation from lot to lot, similar to that occurring in black-and-white 


materials. With fresh film, the variation in sensitivity from one lot 
to another appears to be less than one of the smallest compensating 
filter steps (CC23, CC33, or CC43). The variation in layer sensi- 
tivity from one layer to another appears to be likewise small and of 
comparable order. For film that has aged slightly (for example, less 
than 6 months old when stored at a constant temperature not over 70 
F), there is a slight loss in speed and a slight change in color balance, 
but the sum total of such variations does not appear to exceed the 
equivalent of the single filter step already mentioned. When film 
has been improperly stored or is old it is likely to be "off balance" by 
more than a single filter step. It is prudent to make an exposure and 
color balance test not only for every emulsion lot of film received, but 
also on every lot received at different times. A log with the results of 
such tests quickly shows up not only errors and variations, but also 
indicates when properly interpreted the magnitude of the exposure 
variation actually encountered in printing the tests. 

Routine testing of filters in the arrangements in which they are used 
is helpful. The cost for several curves is no more than the price of a 
single 800-f t roll of raw film. Testing of this kind is in reality inex- 
pensive quality insurance. 

Routine testing of the duplicates made commercially is readily 
accomplished if a test strip is attached to every roll of preprint ma- 
terial and printed as a part of the routine printing operation. The 
test strip can be designed to fit whatever situation is required. A 
simple yet informative test leader might include a few frames of each 
of the following: 

(1) A fine-grain silver film (such as EK 5365} uniformly exposed to a density of 

(2) Clear leader (made by running EK 5365 through hypo) . 

(3) The 16-mm Kodak color test chart. 

(4) Kodachrome printed through a Wratten 49 (blue filter) to yield a color 
density of 1.0. 

(5) Same as (4) except through a Wratten 61 (green filter). 

(6) Same as (4) except through a Wratten 29 (red filter). 

(7) A resolving power test chart. 

The test strip should be printed at "normal" exposure for the particu- 
lar lot of raw film. 

Sections 1, 2, 4, 5, and 6 can be read with an ordinary Capstaff- 
Purdy (EK Co.) densitometer. The results of these readings can be 
plotted in a "scatter diagram" in accordance with the methods de- 

132 W. H. OFFENHAUSER, JR. Vol 45, No. 2 

scribed in ASA Standards for Quality Control, Zl.l, Z1.2, andZl.3. 
A check lasting over even a few weeks will indicate where the "tight- 
ening-up" process should be applied for quality improvement and in 
what amounts it should be applied. If more accurate results are 
required in specific cases, other checks such as per cent-transmittance 
versus wavelength curves may be taken on the pertinent parts of the 
printed strip. 

Conclusion. Integral tripack films such as Kodachrome are good 
materials in good control at the present time. It must be remem- 
bered, however, that there is no "perfect" color process. The usual 
requirements of a color process are : 

(7) A suitable gray scale, 

(2) Comparable color scales for the components, 

(3) Accurate reproduction of color, 

(4) Good differentiation of color. 

Each of these requirements conflicts in some measure with at least one 
of the other three. Ordinarily (1) and (4) are favored over (3). 

The most recent edition of "Photomicrography" (1944) carries the 
admonition concerning photometric filters in these words, "There are 
no colored pigments or dyes actually available for making the three 
colored components of a picture that do not absorb light outside of 
their own spectral domain and thus degrade the hues of the final 
result". 9 Curves are essential in describing the performance of filters 
for color duplicating purposes. 

For convenience, wavelengths shorter than approximately 420 m/* 
are filtered out of the exposing illumination. This is accomplished 
with the Wratten 2 A filter or equivalent. The near ultraviolet and 
the shorter blue rays increase the exposure in the blue layer, but as it 
is not practicable to evaluate this exposure accurately by either an 
illuminometer or by the indirect film method, it is usually better to 
remove these wave lengths from the light beam. With these wave- 
lengths removed, results are usually more reproducible. 

Generally speaking, colors that have components in the filter cross- 
over regions of 500 m/z and 600 m^t are difficult to control for accuracy 
of color reproduction. In most practical cases good color differenti- 
ation will suffice and such filters as the Corning 51 20 (light didyrn- 
ium) and its approximate equivalent, Jena BG11, in moderate thick- 
nesses, such as 1 mm and 2 mm, are often quite useful. These filters 
are likewise helpful in retaining face and skin detail in a duplicate that 


is printed from a slightly overexposed original, as well as in the re- 
production of certain biological stains and dyes 9 whose absorption 
points are "unfortunately" located. 

It should be remembered that even a "perfect" copy is of little value 
if the print is not projected properly. The importance of correct 
illumination level can hardly be overemphasized. At present, even 
when screen illumination is in the uppermost range found in practice, 
it is rarely within 20 per cent of what might be called the optimum 
value. Because screen illumination is of such a low general -order, 
film laboratories have often deliberately chosen to overexpose dupli- 
cates in printing, thereby wiping out much detail and further aggra- 
vating an already serious condition of low resolving power. The 
magnitude of this effect can be roughly judged by comparing the 
detail and quality of a Technicolor picture projected in a neighborhood 
theater with the usual 16-mm projection of a Kodachrome duplicate. 
Such a comparison is reasonable; several Technicolor releases have 
been made from 16-mm Kodachrome originals. There are other fac- 
tors, but these are beyond the scope of this paper. 

Although the Eastman Kodak recommended method of duplicating 
Kodachrome is relatively simple and places the major part of the 
control burden upon the film manufacturer, it is imperative that the 
commercial laboratory accept its share of the control responsibility 
knowingly and willingly, and appreciate the importance of process 
control by applying some of its basic principles. The control of 
emulsion quality in manufacture and the control of color developing 
have reached such a high point that it is no longer possible to indis- 
criminately "pass the buck" to the film manufacturer if prints do not 
come up to expectations. If the control required for good color 
duplication according to today's standards were applied to the release 
printing of the mass-produced prints of Armed Forces training films, 
the result would be beyond our fondest dreams. This, however, is 
putting the cart before the horse, for we must first learn to control the 
single parameter of good monochrome successfully before we can 
expect to be successful with the 3 parameters of integral tripack 
duplication. This is a challenge to all concerned ; a rapidly growing 
industry will be awarded as the prize to those who produce the best 
product at the lowest price. 


1 HOMOLKA, B.: Brit. J. Photog., 54 (1907), pp. 136, 196, 216. 

2 LEWY, E.: German Patent No. 250,647 (1909). 


3 FISCHER, R.: Brit. J. Photog., 60 (1913), p. 595; English Patent No. 15,055 

4 BRUNO, M.: "Maps on Microfilm," /. Soc. Mot. Pict. Eng., XLI, 5 (Nov., 
1943), p. 423. 

6 GORISCH, R., AND GORLICH, P.: "Reproduction of Color Film Sound Rec- 
ords," /. Soc. Mot. Pict. Eng., 43, 3 (Sept., 1944), p. 206. 

6 FRIEDMAN, J. S.: "Monopack Processes," /. Soc. Mot. Pict. Eng., 42, 5 
(May, 1944), p. 274. 

7 GAGE, H. P.: "Color Theories and the Inter-Society Council," /. Soc. Mot. 
Pict. Eng., XXXV, 4 (Oct., 1940), p. 361. 

8 GENERAL ELECTRIC Co. : "Mazda Lamps Bulletin LD-1," Second Printing 
(Sept., 1940), p. 38 (Cleveland, Ohio). 

9 EASTMAN KODAK Co. : "Photomicrography," (1944 Edition), p. 151 (Roches- 
ter, N. Y.). 





Summary. This report details the progress to date in the improvement of the 
release print quality of 35-mm features reduced to 16-mm, which has resulted from 
the work of the Academy Research Council Committee on Rerecording Methods for 
16-Mm Release of 36-Mm Features. It also describes the new 16-mm test film for 
field checking projector adjustment prepared by the Research Council in accordance 
with American War Standard Specification Z52. 2-1944. 

The Research Council of the Academy of Motion Picture Arts and 
Sciences organized its Committee on Rerecording Methods for 16-Mm 
Release of 35-Mm Features on February 17, 1944, with Wesley C. 
Miller as chairman. The committee was assigned first, to investigate 
the procedures then used in making 16-mm release prints from 35-mm 
originals and to determine the characteristics of the equipment used 
in reproducing these prints, and second, to effect an improvement in 
16-mm release print quality. 

In investigating procedures and equipment then used, excerpts 
from a number of 16-mm release prints were seen and heard on 
several makes of projectors and on different projectors of the same 
make. It was very obvious that the quality, especially the sound, 
was extremely poor and very often not commercial. In many cases 
a great part of the dialogue, especially low-level passages, could not 
be understood or even heard. 

The poor sound quality was attributable to 3 factors; first, inade- 
quate reproducing equipment; second, inadequate processing facili- 
ties; and third, a release print characteristic designed for 35-mm 

Compared to 35-mm equipment, present 16-mm equipment has 
poor film motion, limited frequency response, insufficient amplifier 
capacity, and horn systems of low efficiency which are inadequate 

* Presented Oct. 17, 1944, at the Technical Conference in New York. 
** Metro-Goldwyn-Mayer Studios, Culver City, Calif. 


136 W. C. MILLER Vol 45, No. 2 

for the purpose of reproducing prints with 35-mm characteristics. 
This is not the fault of the equipment as it is being used under condi- 
tions and for purposes for which it was not designed. Existing 16- 
mm equipment has been designed primarily for amateur or semi- 
professional use with small audiences, and to project prints with a 
comparatively narrow volume range. However, it is being used in 
large auditoriums and with the projector in the auditorium but with- 
out facilities to reduce projector noise. As a result, auditorium noise 
level is high and the volume range which the equipment can handle is 
thereby restricted still further. However, this equipment was in the 
field and in use and could not be improved or replaced immediately. 
This condition will be improved when projectors in accordance with 
the American War Standard Specification for Class I Service Model 
16-Mm Sound Motion Picture Projection Equipment, Z52. 1-1944, 
reach the field. 

Laboratory equipment and procedures did not keep pace with 
requirements for obvious reasons. Sufficient new equipment could 
not be purchased under war conditions. Test equipment was often 
on priority and not available. Present procedures were set up under 
adverse conditions because of the tremendously increased use of the 
16-mm film. Sufficient time, materials, and manpower were not 
available for proper study and coordination. Since that time, this 
phase of the work has been coordinated by the ASA War Committee 
on Photography and Cinematography and improvements have been, 
and are still being, made. 

At that time the general procedure in obtaining a 16-mm release 
print sound track was to make a photographic dupe of the 35-mm 
release print sound track. Thus, subject to the printing and proc- 
essing procedure and the limitations of the reproducing equipment, 
the release characteristic was based on the 35-mm release and had a 
comparable volume range and frequency characteristic. 

As it was not feasible to replace existing 16-mm reproducing equip- 
ment, or make immediate improvements in the laboratory situation, 
it was decided as a temporary expedient to fit the release print to the 
reproducing equipment. This meant an apparent reduction in qual- 
ity of the sound track as it existed on the release print but a net im- 
provement in the reproduced quality, to be provided by reducing the 
volume range and altering the frequency characteristic of the 16-mm 
release track. 

It might seem that the obvious thing to do was to set up a standard 

Aug., 1945 16-MM RELEASE OF 35-MM FEATURES 137 

recording characteristic based on the characteristic of 16-mm repro- 
ducing equipment. However, this is not practicable. It would not 
be feasible even if a standard reproducing characteristic could be 

There are a great number of variables affecting the recording char- 
acteristic. It must often be adjusted to the voice characteristic of 
the actor or actress. It depends upon the type of production being 
recorded, set acoustics, the type of microphone, and before the release 
track is made, on the volume range used in the production and on the 
effect desired by the producer. Various methods are therefore used 
depending on equipment and technique, to arrive at a release charac- 
teristic for the theater. It should be pointed out that in 35-mm work, 
although various methods are used, and although no standard re- 
cording characteristic has been established, all product released to the 
theater is commercial and the quality has improved steadily over the 
past few years. 

An additional factor influencing this decision of the Committee was 
the fact that the group had under consideration 16-mm release 
material from original 35 mm. This original 35 mm is made for 
commercial distribution. Obviously, it was not possible to alter the 
recording or release characteristic of the 35-mm product to adapt it 
to the limitations currently existing in the 16-mm field. 

Listening tests on different types of 16-mm equipment and on 
different projectors of the same make demonstrated that the charac- 
teristics of the amplifiers and speakers varied greatly and gave widely 
different results. The use of tone controls also added a variable in 
the reproducing characteristic. 

As a result, the Committee, with all of the above variables in mind, 
set out to determine upon a rerecording characteristic. 

As all of the studio group are familiar with the Research Council's 
Theatre Sound Test Reel, this reel was used as a basis for 16-mm test 
material. For those not familiar with this reel, it contains excerpts 
from feature release prints from all of the major studios, each sample 
being processed by that studio's laboratory and these samples spliced 
together to make the final complete reel. It thus provides a test film 
exemplifying release quality. This film is used to conduct listening 
tests in auditoriums and to adjust the equipment to optimum condi- 

Using a 35-mm print of the Theatre Sound Test Reel, the following 
16-mm test tracks were made as indicated: 

138 W. C. MILLER Vol 45, No. 2 

(1) A 16-mm photographic dupe. 

(2) A 35-mm variable-density negative rerecorded from the release print sound 
track. This special negative was used to make a 16-mm print by the optical 
reduction method. 

(5) A 35-mm variable-area negative rerecorded from the release print sound 
track. A 16-mm optical reduction print was made from this negative. 

(4) The 35-mm release track rerecorded to a 16-mm variable-density negative. 
A 16-mm contact print was made from this 16-mm negative. 

(5} The 35-mm release track rerecorded to a 16-mm variable-area negative from 
which was made a contact print. 

These tests provided a 16-mm print made as a photographic dupe, 2 
variable-density prints containing material originally made as 35-mm 
variable density and variable area, and 2 16-mm variable-area prints 
from material originally 35-mm density and area. The same 35-mm 
print was used in each case as a rerecording print. 

All rerecordings were made with a reduced volume range and a re- 
stricting frequency characteristic. In general, the frequency charac- 
teristic was sloping on the low end with the low-frequency cutoff in 
the neighborhood of 100 cycles. The characteristic was peaked in 
the intelligibility range and the high-frequency cutoff in the neighbor- 
hood of 5500 cycles. 

After these tests had been prepared, several meetings were held and 
listening tests conducted in different auditoriums as well as out of 
doors, again using several different 16-mm projectors. As a result of 
this second group of listening tests, a rerecording characteristic was 
tentatively decided upon and additional test recordings made. These 
test rerecordings were made on the basis of this rerecording character- 
istic, altered slightly depending upon the type of product being re- 

Following additional listening tests, recommendations were made 
by the Research Council to the producing studios as follows : 

(1} That an especially rerecorded negative be provided for all features and short 
subjects produced for 35-mm release but reduced to 16-mm for distribution over- 

(2) That the volume range of the 16-mm release should be compressed to 
approximately 15 db on the basis that the present 35-mm release print volume 
range is approximately 35 db. This means that main title and montage music 
and dialogue peaks should be rerecorded at full modulation. Any signal re- 
recorded at a level of more than 15 db below full modulation will probably not be 
audible when reproduced except under the most favorable conditions. 

(5) That the following rerecording characteristic be used as a guide (Fig. 1) : 
Low-frequency cutoff 100 cycles; high-frequency cutoff 5000 cycles; a re- 

Aug., 1945 



duced response below 1000 cycles ; and an increased response of approximately 3 
db at 3500 cycles. The resulting characteristic slopes smoothly from 3 db up at 
3500 to approximately 6 db down at 200 cycles. 

(4) That for 16-mm variable-area release the rerecording be either to 35-mm or 
directly to 16-mm, but whichever process is employed separate sound and picture 
negatives be used in making the 16-mm release print. Separate film is recom- 
mended as the picture and sound negative require different processing for the best 

(5) That for 16-mm variable-density release the rerecording be to a 35-mm 
sound negative. From this negative and through the optical reduction process is 
made a 16-mm release print sound track. At the time this recommendation was 

FIG. 1 

made recording and laboratory equipment and facilities were not available to pro- 
vide other than test rerecordings directly to 16-mm. 

(6) That the rerecording be done from the regular domestic release print sound 
track as an expedient to produce a reasonably good result at a low cost. 

The above recommendations were made in the nature of a progress 
report as the rerecording characteristic recommended and the 
methods employed were of a temporary nature subject immediately 
to any improvements in projection equipment and adopted to effect 
an immediate improvement in 16-mm feature release quality. 

The rerecording characteristic was established only as a rough 
approximation of the slope of the equalizer curves and the cutoffs 
recommended. In recommending this rerecording characteristic, 

140 W. C. MILLER Vol 45, No. 2 

the Committee emphasized that it was approximate and temporary 
and its use depended upon the type of product being rerecorded. 
Listening tests had demonstrated that different types of product re- 
quire different treatment. 

Since these recommendations were made, the same rerecording 
characteristic is being used subject to any modification necessary 
considering the type of material being rerecorded. For variable 
density the domestic release print sound track is being used as re- 
recording print to make a special rerecorded negative. A print of 
this negative is optically reduced to furnish a 16-mm negative from 
which contact prints are made. For variable-area release the domes- 
tic release print sound track is rerecorded directly to a 16-mm nega- 
tive from which contact prints are made. 

During the time this work was going on, both the Research Council 
and the Society of Motion Picture Engineers were represented on the 
ASA War Committee on Photography and Cinematography. As a 
part of this work, the Society took on the job of making available 
many types of 16-mm test films. The Research Council was asked to 
make available a 16-mm test film comparable to our 35-mm Theater 
Sound Test Reel. [A print of this 16-mm reel was projected after the 
paper was read Ed.] This reel consists of an explanatory title with 
main title music, excerpts from feature productions released on 
variable-density track, and a piano recording and excerpts released on 
variable-area track. This test film is known as the "16-Mm Test 
Film for Checking Projector Adjustment" and has been prepared by 
the Research Council in accordance with American War Standard 
Specification, Z52.2-1944. The specification was prepared in draft 
form by the Council at the request of Z52 Subcommittee B on 16-Mm 
Sound and was recommended for approval as an American War Stand- 
ard by that Subcommittee with only minor editorial changes. The 
specification for the test film was promulgated as an American 
War Standard on October 11, 1944. 

The variable-density track was made as follows: The 35-mm re- 
lease print sound track was used as a rerecording print to make a 35- 
mm variable-density negative. In the rerecording process the 
volume range was reduced and the frequency range restricted in 
accordance with the rerecording characteristic given above. A com- 
pressor was used in the circuit to automatically control peaks with the 
amount of compression determined by the material in each sample. 
Previous tests had demonstrated that manual control of the volume 

Aug., 1945 16-MM RELEASE OF 35-MM FEATURES 141 

range was not too satisfactory. In rerecording for 16-mm feature 
release, a compressor may or may not be used in the rerecording cir- 
cuit, depending upon the type of product and the technique used. 

A print of this 35-mm rerecorded variable-density negative was 
optically reduced to provide a 16-mm negative from which contact 
prints were made. 

The variable-area release print tracks were rerecorded directly to a 
16-mm negative, again employing a reduced volume range and a re- 
stricted frequency range and using a compressor in the rerecording 
circuit. This negative was used to make 16-mm contact prints. 

Thus the 16-mm sound release negative from which these release 
prints are made consists of a variable-density negative from original 
35-mm variable-density and a 16-mm variable-area negative made 
from original 35-mm variable-area material. 

We would like to point out that the Committee deliberately chose 
this rerecording characteristic with full knowledge that it is possible 
to obtain better sound quality with a different characteristic if proj- 
ecting on good 16-mm equipment in a room with reasonably good 
acoustics. Under such conditions more low-end would be used and 
the volume range increased. However, it has been found that quite a 
few projectors are being used with speakers .having a low-frequency 
resonance which tends to mask out the intelligibility. Tone controls 
are not always used to the best advantage and many instances have 
been encountered where the tone control has been set to cut out the 
highs again reducing the intelligibility. Projection conditions are 
very often adverse as the equipment may be used out of doors or in 
rooms with poor acoustics and high noise level. Consequently, we 
believe that the characteristic decided upon is safer than the charac- 
teristic with more low-end although in some instances and under 
favorable conditions the quality would be improved. A wider volume 
range under good projection conditions would, of course, be preferable 
to the volume-range of the projector test film, but it has been the sole 
purpose of the Committee to arrive at the characteristic giving the 
maximum amount of intelligibility throughout a picture so that 
the dialogue can be heard and will be intelligible under all conditions. 

Accompanying each print of this test film is an instruction sheet 
briefly describing the use of the reel. This instruction sheet includes 
a general description of the contents of the film, the procedure in using 
the film, and the manner in determining the cause of operating 
troubles if they exist. 

142 W. C. MILLER 

This projector test reel is marked Issue No. 1 as it is expected that 
the rerecording characteristic will be revised both to improve the reel 
and to keep in step with advancements in the 16-mm field. 

Since the Committee was formed, in February, 1944, all of the 
producers of feature releases have agreed to rerecord for 16-mm re- 
leases where necessary. One studio was rerecording at 'the time the 
Committee was formed. At least 3 others were rerecording within 90 
days after the formation of the Committee. Thus, since July, 1944, 
a very definite improvement has been made in the release print sound 
quality of a large amount of feature product which has been released 
on 16-mm for distribution overseas. Although the Committee's work 
is not finished, it has accomplished practical results in a relatively 
short time in bringing about this improvement in release quality and 
in making available a 16-mm projector test film. 

[Ed. Note. Since the presentation of this Report, and during the past few months, 
equipment has become available for rerecording directly from the 35-mm version to a 
16-mm release negative. Some recent pictures have been released by this method and 
it is probable that this will ultimately become the standard method as it eliminates any 
irregularities which may occur during the process of optically reducing from 35-mm to 
16-mm sound tracks. 

Also, it should be noted that the Report applies primarily to black-and-white re- 
leases. A similar study on color 16-mm releases has been under way although it 
was somewhat delayed on account of the different problems presented by the color 
processes. A full report on this phase of 16-mm release work is contemplated as soon 
as practicable.] 



Summary. The maximum force between a cam and its follower depends on the 
mass of the driven elements and the maximum accelerations imparted to them by the 
cam. The magnitudes of these forces affect the wear on the surfaces and the tendency 
of the mechanism to be noisy. If the follower works on one side of the cam only, con- 
tact being maintained by a spring, the required spring tension is determined by the 
acceleration. Some cams can be designed to give a predetermined motion and if the 
motion can be simply expressed mathematically, the acceleration can be easily calcu- 
lated. Other types of cam are made up of a series of circular arcs and the motion of the 
follower is determined by certain geometrical relations. 

For any given design, the position of the follower can be calculated point by point 
of the cam rotation, by solving triangles, or graphically. But to determine velocities 
by measuring slopes of the position curve, and accelerations by measuring slopes of the 
derived velocity curve, gives only rough approximations. Determination of velocity by 
writing and then differentiating a mathematical equation for follower position in terms 
of cam position, appears to be possible only for the simplest case, but if the problem is 
taken in 2 steps, first to find the follower position by the solution of one or 2 triangles, 
and then for a given position, to find the velocity and acceleration from formulas given 
in the paper, the calculations are not difficult, and the degree of precision may be 
whatever is required. 

Anyone who has tried measuring slopes of plotted curves for such 
purposes as determining velocities or accelerations realizes that the 
method gives very crude results. In the course of a study of claw mo- 
tion for projector pull-downs it was desirable to find whether a cer- 
tain arrangement which shortens the pull-down time resulted in ob- 
jectionably large forces against the cam. The graphical method 
yielded answers which looked unreasonable, and a more accurate 
method was sought. 

The reader might appropriately ask at this point whether a cam 
can not be designed to give almost any desired relation between cam 
and follower positions, and can not the movement curve be chosen in 
accordance with some simple mathematical law for which the accelera- 

* Presented Oct. 18, 1944, at the Technical Conference hi New York. 
** RCA Victor Division, Radio Corporation of America, Indianapolis, Ind. 




Vol 45, No. 2 

tions are known to be within specified limits? This is true in general 
of cams with cylindrical or roller followers. The cam is laid out by 
making a series of drawings to scale, each of which shows the cam fol- 
lower in a new position relative to the cam, and the cam surface is the 
envelope of the series of arcs which represent the successive outlines of 
the follower. This is illustrated in Fig. 1. 

||On the other hand, there are important types of cam in which 
the designer has less freedom. For example, in the Geneva movement 
(which in principle can be considered to be a cam) the movement 

FIG. 1. Method of laying out cam with roller follower to give 20-degree 
swing of follower arm in 90 degrees of cam rotation sine wave movement. 

of the star- wheel is completely determined by the spacing of the shafts 
and the radius to the pin. Another important type is the constant- 
diameter cam, illustrated in its simplest form in Fig. 2. This employs 
a flat follower, which is an advantage on the score of simplicity and 
distributed wear. The constant-diameter feature permits operating 
the cam between 2 parallel flat surfaces, thus giving positive drive 
at all times and making use of a spring unnecessary. If the cam fol- 
lower does not swing, but maintains the sliding surfaces in a fixed di- 
rection for example, if the follower slides on a rail its motion can 
be very simply stated. Motion begins when, the line O-P\ passes the 
vertical (x = 90 degrees), and thereafter the upper follower surface 



FIG. 2. 

Constant-diameter cam in its 
elementary form. 

moves down with PI, giving a displacement from the starting position 
equal to ^4(1 sin x) until line D becomes vertical, after which the 
upper slide rests on the arc S\- 
Sz, whose center is PZ, and which 
is D above P$. The motion now 
depends on the movements of 
P 2 , the second half of the curve 
being identical in shape with the 
first half. The displacement, 
velocity, and acceleration curves 
for the case of A = 1 and 8 = 
60 degrees are shown in Fig. 3. 
It will be noted that the accelera- 
tion is a small portion of a sine 
curve and then reverses suddenly to equal values of retardation. 
It is obviously not desirable to make the cam with sharp corners. 
Hence each radius is increased by a small addendum r, giving a small 

arc of radius r at the previously 
sharp corners, as shown in Fig. 
4. This increases the over-all 
diameter of the cam by 2r, but 
with a parallel-motion follower 
does not alter the follower mo- 
tion at all, and with a pivoted 
follower the motion is the same, 
provided the follower surface is 
displaced so that if extended, 
instead of passing through the 
pivot point, it would pass above 
it by the same distance that the 
cam surface is above P, namely 
by r. 

In most applications of the 
constant-diameter cam the fol- 
lower is pivoted at a short dis- 
tance from the cam shaft, with 
the result that there is simultan- 
eous rotation of both cam and follower. This has the effect of 
shortening the time of one movement of the follower and lengthen- 
ing the time of the reverse movement. For example, if the action 

















90 60 70 60 50 40 30 

FIG. 3. Characteristics of cam 
shown in Fig. 2. or 4, with A = 1 ve- 
locity and acceleration for cam rotating 
one radian per sec. 



Vol 45, No. 2 

angle 6 (in Fig. 2 or 4) is 60 degrees, and if the follower makes a swing 
of 5 degrees, it will execute one movement in 55 degrees of cam 
rotation and the reverse movement in 65 degrees. Advantage may 
be taken of this relation to shorten the pull-down time. The required 
diameter of the cam for a given throw (difference between the larger 
and the smaller radius) varies approximately as the inverse square of 
the action angle 6, or in other words, quickening the pull-down means 
enlarging the cam. Hence there is generally good design reason for 
shortening the pull-down time by the expedient just described, but it 
is desirable to ascertain whether this is being done at the cost of un- 
desirably large acceleration 

If the follower pivot is to the 
right of the cam, for quick pull- 
down the cam would rotate 
clockwise (i. e., x decreasing with 
time) and in deriving formulas 
the increments dx and dy are 
here drawn in this direction, 
and marked negative for mathe- 


dy . 

consistency; is 


FIG. 4. Constant-diameter cam with 
addendum to give rounded corners. 

rate of rotation of the follower 
compared with that of the cam 
(which is in general constant). 
To simplify the problem the cam is assumed to be as shown in Fig. 
2 (i. e.,r = 0) . During the first half of the movement the flat follower 
B pivoted at M, Fig. 5, rests on the corner PI which moves in an arc 
of a circle of radius A , about cam shaft center O. The determination of 
y requires only the solution of a triangle (see Fig. 6), with 2 sides 
(radius A and center distance C) and the included angle x given. 
The simplest procedure is to assume z, calculate y from the formula 
that the sines of the angles are in the same ratios as the sides opposite 


sin 2 = sin (x + y) 

A Smy 

and from the relation that z = x + y, find x. 

We can differentiate both sides of Eq (1) with respect to x and then 

solve for , but the geometrical approach now to be explained gives it 



in simpler form, and is moreover applicable to the second case, or part 
of the problem, which is more complicated. 

For any given cam position (designated by x) if we know the cor- 
responding value of z (or x + y) and y, it is quite a simple matter to 
find the rate of change of y when x changes at a specified rate. Refer- 
ring to Fig. 6, if x changes by a very small angular negative increment 
dx (expressed in radians), the point PI will move a distance Adx 
in a direction perpendicular to A. Let this movement of PI be 

FIG. 5. During the time that follower B is resting on 
corner P^ the relation between x and y is determined by 
solution of a triangle with 2 sides A and C constant. 

divided into 2 components Adx sin z and Adx cos z, parallel and 
perpendicular to B, respectively. It is evident that the part of the 
movement of PI in line with B will not change y, while the perpen- 
dicular component Adx cos z will decrease y by the angle 

dy = - . In the limit it makes no difference whether we 

divide by B or by B Adx sin z. Hence 





which gives the ratio of angular speed of the follower to that of the 



Vol 45 No. 2 

To find the rate of change of follower angular speed we differentiate 
with respect to x. Both z and B are variables depending on x. 

and differentiating the expression as a fraction 

d . dB 




Referring to Fig. 6, 

= sin 2 ( 1 -H -g cos 2 J 


FIG. 6. Components of Adx parallel and perpendicular to B. 

bB = A dx sin 2 


-=- = A sm 2 



Then substituting (4) and (5) in (3) 


( 1 + ^ cos z } 
\ n J 

sin 2 ( 1 + ^ cos 

A sin 2 cos 2 

A A* 

-17- sin 2 2 -^ sin 2 cos 2 

From a consideration of Fig. 5, it will be evident that the problem 
just considered is applicable to the case of a Geneva movement, the 
point Pi representing the center of the pin, and the line B correspond- 
ing to the center of the slot in the star- wheel. For the usual 90-degree 
Geneva movement x varies from +45 to 45 and C = V2A. Fig. 7 
shows the calculated star-wheel velocities and accelerations, referred 
to the angular velocity of the pin wheel. For any given value of the 

pin-wheel speed n revolutions per sec, should be multiplied by 2irn 





to give star-wheel angular velocity in radians per sec, and multi- 

dx z 

plied by (Sirri) 2 to give angular accelerations. 

We have so far dealt with only the simpler part of the calculation 
of motion in the case of the constant-diameter cam with pivoted fol- 
lower. For half the stroke the follower rests on the corner, Fig. 2 (or on 
a small-radius arc just outside it, the error in assuming contact at the 
corner being small unless r is larger in relation to C than is ordinarily 
the case) . After the line D, in Fig. 2, passes the position perpendicu- 

40 30* 20 10 -10 -20 -30 -4C 

FIG. 7. Displacement, velocity, and accelera- 
tion of star- wheel. 

lar to follower B, the latter no longer rests on the corner but on the 
arc Si-S 2 of which P 2 is the center, as indicated in Fig. 8. Thereafter 
the calculation of y requires the solution of 2 triangles, OMP Z in Fig. 
8, to find u and E, and then the right triangle P 2 TM to find u + y and 
B. It is scarcely necessary to include here the steps of calculating y. 

During this part of the throw, the changes in y depend on the mo- 
tion of P 2 , and the effect on y of a small movement Adx of P 2 de- 
pends on the components of Adx parallel and perpendicular to 
B. We must therefore figure out the angle between the direction of 
motion of P 2 and follower surface B, and this is the same as that be- 
tween A and R, indicated as Win Fig. 9, in which it is seen that 



Vol 45, No. 2 

-9Q degrees 


If x decreases by dx, PZ moves Adx cos W parallel to B and 
Adx sin W perpendicular to B, as illustrated in Fig. 10A. Move- 
ment of P 2 parallel to B (a right angle being maintained at T, and R 
being constant) simply lengthens B without affecting y. If P 2 moves 
downward by Adx sin W, it pulls T down, by the same amount, or 

, Adx sin W , 
decreases y by , whence 



FIG. 8. While follower B rests on arc SiS t , finding y for 
a given x calls for solution of 2 triangles, first MOPz and 
then MTP t . 

Then, sin W and B being both functions of x, 

3? "A 

B~ sinW - sinW 
dx dx 


In this we must substitute the values of sin W and 

dx dx 

- sin W = cos W, and from (7) and (8), 
dx dx 

, whence 

- sin W = cos wl + sin 



Both components of Adx affect B. As just stated Adx cos 
W lengthens B by the same amount, while Adx sin W, in changing 
y, also changes the direction of R by the same angle, namely, dy 
(since 'R and B are perpendicular to each other) . As shown in Fig. 

~ sin 

Wdx). Then, adding the effects on B of the 2 components of Adx, 
dB = -Adx cos W - #4 sin Wdx 


dB Tr . RA . ,,, ni\ 

j = A cos W 5- sin W (II) 

ax Jo 

W = a -f-X 1-0-90*= Angle between A own 4 3 

FIG. 9. Establishment of angle between A and .R. 

Substituting Eqs (10) and (11) in (9) 
^ = ~^B cos.w(l + ^ sin W^ - sin W (- 4 cos W - ^ si 

A A A A% T) A 2 

= ^ cos T^ + ^ cos T^^ sin PT + ^5 sin W cos W + - sin 2 
Jo > JD JD* 

= cos PF + 2 cos JF sin w + sin 

The above expression does not appear to be written in its simplest 
form, but for purposes of calculation, it is believed to be in the most 
convenient form. 

Fig. 11 shows the calculated displacement, velocity, and accelera- 
tion curves for a flat follower held against a constant-diameter cam, 
with center distance C equal to 2A . The 2 parts of the curve are not 



Vol 45, No. 2 

identical, as was to be expected with different formulas, but the dif- 
ference is not large. 

The method of calculating the second part of the curve (Eqs (8) 
and (12)) may be applied to the problem of making allowance for the 
addendum r at the corner, for it covers the general case of a flat fol- 
lower passing through a pivot point M, and maintained tangent to a 
circular arc whose center moves in a circle about the cam shaft axis O. 

FIG. 10A. Components of Adx, parallel and perpendicular to B. 
FIG. 10B. Effect on B and y of movement of P 2 in direction of 
perpendicular to B. 

The expression for the angle W will have to be correctly formulated in 
each case. 

If the surface of the follower is not radial with respect to its pivot 
the formulas (8) and (12) may be used by choosing an appropriate new 
value for R. This may be illustrated for the common case in which 
the follower is in the form of an open rectangle whose mid-line MN, 
Fig. 12, passes through the pivot axis. The total diameter of the cam 
is D + 2r and the 2 follower surfaces are this distance apart. During 
the first part of the downward movement, the surface .Fi is tangent to 
the arc S 3 Si, which is r above PI. At the same time the mid-line MN, 



which is 4- r below FI, is - below PI and therefore tangent to a 
2 2 

circular arc S^e, with P\ as the center and as its radius. When the 


point of tangency of FI passes to the arc SiS 2 , motion is determined 
by the position of P 2 . F\ is D + r above P 2 measured perpendicular to 

FI. At the same time the mid-line M N is above PI and therefore 







FIG. 11. 





) AND 


r EC 






25 3> 








d 2 y 





















' d7 


I N 


65- 60 55 50 45 40 35 30 25 20 15 10 5 

Motion characteristics of cam of Fig. 2, with pivoted follower, 
C = 2A. 

tangent to a circular arc 

having its center at P 2 and a radius *-* 


In the case of the follower whose surfaces FI and F% are parallel and 
equally distant from the radial line MN, the 2 parts of the movement 
are symmetrical, and therefore only one part needs to be calculated. 

Some comments on the several curves of acceleration may be in 
order : If a mass is to be moved through a given distance in a speci- 
fied time, this can be accomplished with the smallest maximum ap- 
plied forces if the acceleration is maintained constant for half the time 
or distance, and then an equal constant retarding force applied until 
the mass is brought to rest. The curve of displacement versus time 



Vol 45, No. 2 

consists of 2 parabolic segments. To accomplish the same travel, sinu- 
soidal movement would require 1 1 per cent longer time with the same 
maximum acceleration. Reference to Fig. 3 will show that for the 
case considered, acceleration and retardation are maintained through- 
out their periods of action between maximum and 0.87 of the same, 
thus approximating very closely the constant acceleration condition. 
Figs. 1 1 and 13 show that with pivoted followers, the acceleration 
is not maximum at the start but increases (in the case illustrated) to a 
maximum of about twice the initial value. 

FIG. 12. Cam with open rectangle follower, an example of the more 
general problem of a follower with a flat surface which is not radial 
with respect to the pivot point M. 

Figs. 11 and 13 were calculated for cases of unusually short relative 
distance between shaft and pivot. A very large distance would give 
characteristics approximating those of Fig. 3 in which the accelera- 
tion falls off slightly from the initial value. With medium pivot dis- 
tances the conditions of constant acceleration and retardation may be 
approximated even more closely than Fig. 3. 

From the standpoint of mechanical wear, constant acceleration or 
an approximation thereto is desirable, but in other respects a lowered 
initial value of acceleration and final value of retardation may have 
some advantages, for this means that the film gets under way more 
slowly and comes to rest more gradually, and these in turn mean re- 
duced travel ghost for the same over-all pull-down time and shutter 
blade width. The Geneva movement, see Fig. 7, is a rather extreme 



example of maximum acceleration near the middle of the stroke, the 
maximum acceleration being about 5.5 times the initial, and it is 
known to be quite common practice to permit considerable light on 
the screen at the instant that the pin enters and leaves the slot. 

Ordinarily dependence is placed on gate friction to apply the neces- 
sary retarding force to the film itself, but assuming that the film is suf- 
ficiently supported between the gate and the sprocket so that it will 
not buckle, it will be recognized that some of the retarding force may 
be provided by the sprocket teeth, provided this does not occur too 
near the end of the movement. In order to afford accurate registra- 
tion or freedom from picture jump, the front face of the tooth must 


90 85 80 75 70 65 60 55 50 45 40 35 30 

FIG. 13. Motion characteristics of cam and follower of Fig. 12. 

be in contact with the front edge of the perforation at the end of the 
pull-down. If at any time it loses contact (as it must do if the sprocket 
helps to retard the film) it must reestablish contact before the end. 
Gate friction must therefore exceed film retardation force, throughout 
a sufficient distance at the end, for the sprocket to catch up, or in 
other words, for a distance somewhat greater than the clearance or 
backlash between tooth and perforation. In the case of the Geneva 
movement, it seems reasonable to believe that at the time of maximum 
retardation, the back of the tooth may engage the back of the perfo- 
ration and thus help retard the film, but the sprocket still has room to 
catch up to the film and end the pull-down with forward tension on 
the film, which is necessary for controlled final position. 


The items appearing in this section were submitted July 12, 1945, by members 
of the Technical News Committee, who welcome and will consider items of current 
technical interest from any member of the Society. 

Additional information concerning these items, or the equipment and processes 
discussed, may be obtained by communicating with the General Office of the 
Society, Hotel Pennsylvania, New York 1, N. Y. 


The Optical Department of RKO Radio Pictures, which is under 
the supervision of Vernon Walker and Lin wood Dunn, has been 
engaged in duping several reels of the old 28-mm film to 35-mm for 
release in the Flicker Flashback series of RKO shorts. The film was 
chosen from a library of over 100 reels owned by Walter Green. 
Twenty-eight millimeter film was the early standard size for non- 
theatrical use on noninflammable stock. This film was popular 
shortly after World War I and was finally supplanted by 16 mm. 

Among the films duped are Chaplin's Adventurer (1916), D. W. 
Griffith's 2-reeler The Battle of Elderbush Gulch with Lillian Gish 
(1913), and Romeo and Juliet with Florence Turner (1910). 

A 28-mm Peerless projector was adapted to the optical printer and 
the results were excellent. The 28-mm film is in good condition 
probably owing to the fact that it was never subjected to the wear of 
theatrical use. 


The framework for commercial post-war television has been very 
nearly completed by the Federal Communications Commission in 
their report of June 27, 1945, and their current correlative activities. 
The final allocation is as follows : 

Freq. Band 

(Me) Proposed Allocation 

42-44: Nongovernment Fixed and Mobile 

44-50 Television Channel No. 1 

50-54 Amateur 

54-60 Television Channel No. 2 



60-66 Television Channel No. 3 

66-72 Television Channel No. 4 

72-76 Nongovernment Fixed and Mobile 

76-82 Television Channel No. 5 

82-88 Television Channel No. 6 

88-92 Noncommercial Education FM 

92-106 FM 

106-108 Facsimile 

An informal technical conference was held by the FCC on July 13, 
1945, to pass upon the standards formulated by the Radio Technical 
Planning Board over the past year of more. With the Radio Manu- 
facturers Association Engineering Committee on television fairly 
well along in formulating the industry practices under the RTPB 
standards, the way is clear for detailed engineering design of both 
broadcasting and receiver equipment. 

Since numerous television broadcasters are already in operation in 
the large cities, the responsibility for rapid progress in commercial 
television lies with the television receiver manufacturer. As soon as 
he can produce in quantity and sell good television merchandise, the 
advertiser will enter the field and the perennial dilemma of who shall 
come first will be solved. Receiver manufacturers will be wise to 
stay with JAN quality components if they wish to do the best for the 
public, the art, and themselves. 

In the Hollywood area an important development has been the 
announcement of the Southern California Edison Company that 
they intend to convert their 50-cycle electric power service to 60 cy- 
cles. This will bring the vast hinterlands of Southern California on 
the same power system as the cities of greater Los Angeles, and 
obviously simplify television broadcasting and receiver technology. 



An interesting discussion of foreign language versions of American motion pic- 
tures was conducted at the meeting of the Pacific Coast Section of the Society on 
June 26. Five speakers from major Hollywood studios took part in the program 
under the title "Symposium on Foreign Language Versions." Those speaking 
and their subjects were: Luigi Luraschi, Paramount, discussed "The Foreign 
Market Past, Present and Future"; John Bodnar, Twentieth Century- Fox, 
spoke on "Methods of Foreign Release of American Pictures"; Jack Cutting, 
Walt Disney Studios, spoke on "Showmanship in Foreign Release of American 
Pictures"; James Stewart, RKO, discussed "Preparation for Foreign Dubbing"; 
and John Livadary, Columbia, spoke on "Recording Techniques for Foreign 

A large audience participated enthusiastically in the discussion period following 
presentation of the papers and exhibition of sample reels of foreign language ver- 
sions. Because of the interest in this general subject, the speakers have been 
requested to prepare their material for publication in the JOURNAL. 

The meeting, which was held in the ERPI Review Room, was preceded by a 
dinner at the Hollywood Athletic Club where the speakers were guests of local 
Section officers. 



Position open for man or woman with experience in optical instrument 
design. Position also open for man or woman with experience in lens 
design or computing. Write for interview. Binswanger and Company, 
Optics Division, 645 Union Ave., Memphis, Tenn. 

Physicist with special training in optics for research on utilization of 
carbon arcs particularly in projection systems. Apply to Research Labo- 
ratory, National Carbon Co., Inc., P. O. Box 6087, Cleveland 1, Ohio. 

Designer and engineer experienced in optics, lighting, and microphotog- 
raphy, capable of designing microfilm reading equipment and products 
related to microfilm industry. Reply to Microstat Corporation, 18 
West 48th St., New York 19, N.Y. 

Design engineer, experienced in mechanics and optics of motion picture 
cameras, projectors, and film scanning. Give details. Reply to Mr. 
John H. Martin, Columbia Broadcasting System, Inc., 485 Madison 
Ave., New York 22, N.Y. 




Sound recording engineer, 16- or 35-mm equipment, studio or location 
work, single or double system. Free to travel. For details write J. J. K., 
354 Ninth Ave., New York 1, N.Y. 

Honorably discharged veteran with 15 years' experience in all phases of 
motion picture production, including film editing, directing, producing. For 
details write F. A., 30-71 34th St., Long Island City 3, N.Y. Telephone 
AStoria 8-0714. 

Projectionist-newsreel editor with 15 years' experience just released 
from service. Willing to locate anywhere. Write P. O. Box 152, Hamp- 
den Station, Baltimore 11, Maryland. 

We are grieved to announce the death of Charles E. Shultz, Active 
member of the Society, on August 8, 1945, at Willever Lake, New 








Vol45 SEPTEMBER, 1945 No. 3 



Some Practical Aspects of the Intel-modulation Test 


8000 Pictures Per Second H. J. SMITH 171 

The Presentation of Technical Developments Before 
Professional Societies W. L. EVERITT 184 

Efficiency of Picture Projection Systems 

E. W. KELLOGG 191 

Technical Problems of Interpretation in Producing 
Foreign- Version Films T. Y. Lo 203 

Some Relationships Between the Physical Properties 
and the Behavior of Motion Picture Film 

R. H. TALBOT 209 

Problems of Theater Television Projection Equipment 


Current Literature 241 

Fifty- Eighth Semi- Annual Technical Conference 243 

Society Announcements 246 

Copyrighted, 1945, by the Society of Motion Picture Engineers, Inc. Permission to republish 
material from the JOURNAL must be obtained in writing from the General Office ot the Society. 
The Society is not responsible for statements of authors or contributors. 

Indexes to the semi-annual volumes of the JOURNAL are published in the June and December 
issues. The contents are also indexed in the Industrial Arts Index available in public libraries. 





Board of Editors 





Officers of the Society 
**President: DONALD E. HYNDMAN, 

350 Madison Ave., New York 17. 
** Past-President: HERBERT GRIFFIN, 

133 E. Santa Anita Ave., Burbank, Calif. 
** Executive Vice-President: LOREN L. RYDER, 

5451 Marathon St., Hollywood 38. 
* Engineering Vice-President: JOHN A. MAURER, 

37-01 31st St., Long Island City 1, N. Y. 
** Editorial Vice-President: ARTHUR C. DOWNES, 

Box 6087, Cleveland 1, Ohio. 
* Financial Vice-President: ARTHUR S. DICKINSON 

28 West 44th St., New York 18. 
**Convention Vice-President: WILLIAM C. KUNZMANN, 

Box 6087, Cleveland 1, Ohio. 
* Secretary: E. ALLAN WILLIFORD, 
230 Park Ave., New York 17. 
*Treasurer: M. R. BOYER, 
350 Fifth Ave., New York 1. 


*FRANK E. CARLSON, Nela Park, Cleveland 12, Ohio. 
**JOHN I. CRABTREE, Kodak Park, Rochester 4, N. Y. 
**CHARLES R. DAILY, 5451 Marathon St., Hollywood 38. 

*EDWARD M. HONAN, 6601 Romaine St., Hollywood 38. 
*tCLYDE R. KEITH, 233 Broadway, New York 7. 

*G. T. LORANCE, 92 Gold St., New York 7. 
**PETER MOLE, 941 N. Sycamore Ave., Hollywood. 
*fHoLLis W. MOYSE, 6656 Santa Monica Blvd., Hollywood. 
** WILLIAM A. MUELLER, 4000 W. Olive Ave., Burbank, Calif. 

*EARL I. SPONABLE, 460 West 54th St., New York 19. 
**REEVE O. STROCK, 111 Eighth Ave., New York 11. 

*WALLACE V. WOLFE, 1016 N. Sycamore St., Hollywood. 

*Term expires December 31, 1945. fChairman, Pacific Coast Section. 
**Term expires December 31, 1946. ^Chairman, Atlantic Coast Section. 

Subscription to nonmembers, $8.00 per annum; to members, $5.00 per annum, included in 
their annual membership dues; single copies, $1.00. A discount on subscription or single copies 
of 15 per cent is allowed to accredited agencies. Order from the Society at address above. 
Published monthly at Easton, Pa., by the Society of Motion Picture Engineers, Inc. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 

General and Editorial Office, Hotel Pennsylvania, New York 1, N. Y. 

Entered as second-class matter January 15, 1930, at the Post Office at Easton, 

Pa., under the Act of March 3, 1879. 


Vol 45 SEPTEMBER, 1945 No. 3 




Summary. The intermodulation test provides a measure of the distortion intro- 
duced by a sound system through its failure to reproduce two or more frequencies with- 
out interaction between them. The principles underlying the measurement of inter- 
modulation are described, and procedures for testing both negative and positive films 
are outlined. Results representative of both standard and experimental stocks and 
processirg are given. 

The interpretation of various families of intermodulation versus print density curves 
in terms of film characteristics is discussed. Test modifications required by the intro- 
duction of noise reduction bias and results obtair.ed under such conditions are de- 
scribed. Application of the intermodulation test to 16-mm films is shown to require 
no fundamental change in the test procedure. 

The intermodulation method of investigating distortion on 
variable-density sound tracks, as described by Frayne and Scoville, 1 
has been of great value in controlling and improving sound quality. 
Limited at first to a comparatively small number of users, interest in 
the method has been constantly increasing. This trend has received 
added impetus from requests by the Armed Forces, directed to the 
American Standards Association, for standards of quality for 16-mm 

In view of this wider interest it appears desirable to describe some 
of the simple and practical features of procedures and conclusions 
which have been encountered during the past few years in using the 
method for the study of emulsions in the Research Division of the du 
Pont Pftoto Products Department. 

* Presented Dec. 13, 1944, at a meeting of the Atlantic Coast Section of the 

** Research Division, Photo Products Department, E. I. du Pont de Nemours 
and Co., Parlin, N. J. 


162 E. MESCHTER Vol 45, No. 3 

The theory and underlying principles of this intermodulation 
measurement, as well as the equipment itself and its applications to 
the study of film processing, have been described by Frayne and 
Scoville 1 and also by Hilliard. 2 No detailed account of the theory 







FIG. 1. Schematic representation of the intermodulation 

will therefore be attempted here, but a brief resume is included for 
the sake of convenience. 

Fig. 1 shows the 2 signals which are usually recorded on the nega- 
tive: one of 60 cycles, and one of 1000 cycles whose amplitude is 25 
per cent that of the 60 -cycle signal. These two are mixed, and re- 
corded so that the combined signal is 2 db below valve clash. Low 


frequencies other than 60 cycles are sometimes employed for special 
work, and 400 cycles has been used as the high frequency. These 
changes, however, require modification of the analyzer circuit and so 
are not easily introduced. 

The negative is developed and printed, and the print is run through 
a reproducer and preamplifier, to wlfich the intermodulation analyzer 
is attached. All of the steps from here on, with the exception of the 
manual adjustment of the input level to the analyzer, are performed 
automatically by the analyzer. 

The complex input wave is first put through a band pass filter which 
removes the 60-cycle component, leaving a 1000-cycle signal. If no 
distortion has been introduced, the amplitude of the 1000-cycle signal 
will be constant as at the left of Fig. 1. If nonlinearity is present, the 
amplitude of the 1000-cycle wave will vary periodically as at the right; 
that is, the 1000-cycle signal has been intermodulated by the 60 
cycles. This variation may be either 60 or 120 cycles, depending on 
whether the distortion is introduced at one or both ends of the density 

The 1000-cycle signal is then amplified and rectified. After filter- 
ing out the 1000-cycle ripple, there remains a d-c signal whose average 
magnitude is indicated as M ; this may or may not have a 60- or 
120-cycle signal superimposed on it, depending on the presence or 
absence of distortion. M is a measure of the amplitude of the original 
1000-cycle signal; manual adjustment brings it always to a constant 
level, which may be designated as 100 per cent. 

The d-c signal, with or without the fluctuation, is put through a 
transformer and again amplified. The steady signal on the left gives 
zero result through the transformer and zero distortion is indicated. 
The alternating component of the signal on the right comes out as in- 
dicated below. The average value* of this signal is a measure of the 
distortion present; after rectification and smoothing it is read by a 
meter calibrated as per cent intermodulation referred to M above as 
100 per cent. These are the bare essentials of the measuring system 
as shown by one measurement. 

The original signals do not need to be derived from any extraordi- 
narily high-quality sources. Commercial 60 cycles is satisfactory for 
the low-frequency component, and a small beat-frequency oscillator 

* Average values are usually used, although peak and rms measurements have 
sometimes been employed. 



Vol 45, No. 3 

can be used to obtain the 1000 cycles. Limits of =*=3 per cent in 
frequency and a total harmonic content of less than 5 per cent should 
be adequate for all but the highest precision work. 

Care should be taken in mixing that the sources do not react with 
each other to give something other than the desired 4 to 1 ampli- 
tude ratio; a cathode-ray oscillograph is a very convenient device 
for checking the wave form of the combined signal. 

Sections 12 to 15 ft in length give ample time for each measurement. 
A few inches of unmodulated and unbiased track should be recorded 
at the start or end of each section for density measurement. 








.H .5 -6 -7 -8 


FIG. 2. Intel-modulation curves showing (A} good intermodula- 
tion, (B) poor intermodulation, and (Q displaced minimum. 

After development each negative section is printed at at least four 
and preferably more printer points, to insure sufficient data to draw 
a good curve. The densities of the unmodulated sections on the final 
print should bracket the usual operating region of 0.55 to 0.60. The 
results are usually represented graphically as shown in Fig. 2, in 
which the per cent intermodulation for each section has been plotted 
against the unmodulated positive track density corresponding to that 

Curve A represents a very good result; minimum distortion is a 
low value at the desired density of 0.6. Curve B is not as good; dis- 


tortion does not reach as low a value as A, but the minimum occurs 
at the proper print density. This curve does not represent bad sound, 
but merely not quite the best. Curve C is a definitely undesirable 
result; to obtain low distortion it has been necessary to print at an 
unmodulated track density of about 0.4 much too light. This dis- 
tortion can arise when a negative intended for UV printing is printed 
with white light; the effective contrast of the positive is then too high 
and it is necessary to work down on the toe of the positive character- 
istic curve to obtain low intermodulation. 











M .5 -6 -7 8 


FIG. 3. Intermodulation curves resulting from use of different 
print stocks; A shows greater latitude than B. 

These curve shapes are typical of intermodulation results in that 
each possesses a minimum, which serves to indicate the one best print 
density for the particular combination of stocks and processing under 
consideration. Intermodulation may actually rise to very high 
values; those in the illustration have been cut off at about 20 per 
cent merely for convenience. 

In Fig. 3 are shown some possible results when the same negative 
is printed on 2 different positive stocks, as before at a variety of posi- 
tive track densities. At the optimum track density, 0.6, no difference 
between the 2 stocks is discernible ; however, for departures from this 
optimum point the distortion of B rises much more rapidly than that 



Vol 45, No. 3 

of A. The printing latitude of A is therefore greater than that of B, 
and A is preferable. The explanation of this effect is fairly compli- 
cated, tracing back to the curvature of the positive characteristic 
curve over the range of operating densities. 

Fig. 4 shows the effect of varying the recorder lamp current and 
hence the negative track density of a standard type of VD negative 
When all these negatives are printed on the same standard positive 
material, the print density required for minimum distortion in each 
case is approximately 0.6, regardless of moderate variations in nega- 






FIG. 4. 

5 fe -7 -8 


Intermodulation results from a negative stock showing 
normal sensitometric curve shape. 

tive density. In making prints, therefore, it is sufficient to work to a 
constant positive track density to insure best results; it is not neces- 
sary to make any allowance for negative density. A negative whose 
characteristic curve is essentially a straight line over a sufficient range 
of density gives this result, although departures can always be ex- 
pected if the recording range is sufficiently great. 

Such a situation does not always exist, as is shown in Fig. 5. In this 
case track density on an experimental negative has been varied by 
changing recorder lamp current and the prints have been made on the 
same standard positive stock mentioned above. Here the positive 
track density at which minimum intermodulation occurs definitely 


does depend on the density of the negative; heavier negatives must 
be printed lighter to obtain best results, while lighter negative tracks 
give lower distortion with darker prints. This situation has interest- 
ing possibilities. It should be possible to use a negative stock of this 
type as a sort of self -timing negative. If all takes are run through at 
a fixed printer point, the tendency will be to correct the print density 
automatically toward lower final distortion. The disadvantage is, of 
course, that the variable print density will give a variable output 




60 .50 

\ \ 


-HO - 


.H -5 .6 -7 


FIG. 5. Intermodulation results from a negative stock having un- 
usual curvature in its sensitometric characteristic. 

The type of negative characteristic curve which produces this result 
is shown in Fig. 6. This long sweeping curve, which is practically all 
toe and with no straight line portion, causes the density swing from A 
to A ' to be quite different from that of B to B ' although the change in 
log E produced by the light valve is the same in each case. Heavier 
exposures use a section of the negative characteristic curve where the 
slope is greater; in order to obtain minimum distortion it is necessary 
to print these farther down on the toe of the positive characteristic 
where the slope is lower. 

The preceding discussion is based upon a signal recorded at 2 db 
below valve clash, nearly full modulation. When noise-reduction sys- 



Vol 45, No. 3 

terns are employed additional complications are introduced and addi- 
tional data are required to insure optimum operation. To obtain 
these data a second section is recorded at 12 db below clash with 10 
db of noise reduction, and the intermodulation is measured and 
plotted in the same way as in the previously described tests. Fig. 7 
shows the 2 db and 12 db curves, in which intermodulation has 
been plotted as a function of the unmodulated and unbiased track 
density. The combination of curves A and C represents a satisfactory 
situation; low distortion at both levels is obtained for an unmodu- 
lated and unbiased track density of 0.6. 









FIG. 6. Sensitometric curve typical of negative stocks 
yielding intermodulation results shown in Fig. 5. 

Curve B results when the recording stock has insufficient toe. 
When the noise reduction starts to operate and the density of the 
negative decreases (to give a darker print and reduce noise) the con- 
trast of the negative remains high. It is then necessary to make a 
lighter than normal positive print, to operate further down on the 
positive toe where the slope is lower. Curves B and C demand that 
the unbiased and unmodulated track density shall be 0.4 and 0.6 
simultaneously to give minimum distortion for both loud and soft 
sounds. In the face of this obvious impossibility it must be concluded 
that the negative stock under consideration (or its processing) is un- 

The application of the intermodulation test to checking of 16-mm 
films and processing is not difficult. The maximum frequency em- 


ployed is 1000 cycles, so it is not necessary to change this on going 
from 35-mm to 16-mm conditions, and no change in the signal gener- 
ating apparatus is required. 

As in the case of 35 mm, it is desirable to check the amplifier of the 
reproducer for possible intermodulation which may be introduced by 
the equipment itself. This may be done by introducing the signals 
directly into the amplifier and noting the reading of the intermodula- 
tion analyzer. 








-2db { -ladb 


of .5 -6 -7 -8 


FIG. 7. Intermodulation curves showing performance in record- 
ing systems using noise reduction. Combination A and C is satis- 
factory while B and C is unsatisfactory. 

The frequency response characteristic for 16-mm amplifiers given 
in the American War Standard for Service Model 16-Mm Projectors 
permits a drop of 9 db at 50 cycles under the output at 1000 cycles. 
This does not interfere with the operation of the test, however, since 
the first step in the analyzer is the filtering out of the 60-cycle com- 
ponent. The signal level is adjusted by observation of the 1000-cycle 
component, so the possible lower response at the lower frequency does 
not in itself introduce any error into the measurement. 

For an over-all check of the reproducing system, including the 
photocell and its coupling circuit, it is necessary to employ a test 


film; specifications for such a test film have not been worked out by 
the American Standards Association at this writing. 

In closing, it should be emphasized that the intermodulation test in 
itself cannot be used to the exclusion of listening tests as an indication 
of sound quality. Each furnishes its own particular type of informa- 
tion, and intelligent use of both in combination should lead to the 
best possible control of stocks and processing. 


1 FRAYNE, J. G., AND SCOVILLE, R. R.: "Analysis and Measurement of Distor- 
tion in Variable-Density Recording," /. Soc. Mot. Pict. Eng., XXXII, 6 (June, 
1939), p. 648. 

2 HILLIARD, J. K. : "Distortion Tests by the Intermodulation Method," Proc. 
1. R. E., 29, 12 (Dec., 1941), p. 614. 


Summary. A brief outline is given of the development and design of the Western 
Electric Fastax high-speed motion picture camera. This camera takes up to 8000 
pictures per sec on 8-mmfilm. 

The engineers and scientists of today have a number of methods at 
their disposal as aids in the study of high-speed phenomena that are 
beyond the range of perception of the human eye. These methods 
include the use of optical lever, shadowgraph, oscillograph, oscillo- 
scope, stroboscope, spark or flash photography, and the high-speed 
motion picture. All of these methods are being used today in ana- 
lyzing high-speed motion pertaining to the war effort. 

The high-speed motion picture camera is the most versatile of 
these devices for studying mechanical motions and other high-speed 
phenomena, in that it may be used independently, or in conjunction 
with some of the above-mentioned devices. It can record phenomena 
regardless of whether the motion is repetitive or nonrepetitive on a 
film that can be viewed either frame by frame for measurement, or as a 
motion picture to portray the action in slow motion. Such pictures 
are a complete and direct representation of the action as compared 
to the indirect or partial indices of motion obtained from such in- 
struments as optical levers, oscilloscopes, etc. The action, when 
photographed at a high rate of speed and projected at a slow rate of 
speed, will be retarded or "magnified" by the ratio of these 2 speeds. 
For example, the newly developed Fastax camera will take pictures 
up to 8000 pictures per sec. When these pictures are projected at a 
rate of 16 pictures per sec, a time "magnification" of 500 to one can 
be obtained. 

In order to take motion pictures at this ultra-high speed, the film 
must be propelled through the camera at a velocity of 100 ft per sec, 
or approximately 70 mph. 

* Presented Oct. 18, 1944, at the Technical Conference in New York. 
** Member Technical Staff, Bell Telephone Laboratories, Inc., New York. 




Vol 45, No. 3 

The conventional start-stop mechanisms of the intermittent-type 
motion picture cameras have been made to operate as high as 400 
pictures per sec 1 although it is unusual for them to run much above 
128 pictures per sec. It would be extremely difficult to build a re- 
ciprocating mechanism to operate reliably at speeds in the order of 
8000 cycles per sec, as would be required in the camera under discus- 
sion, and even if a suitable intermittent mechanism could be devised, 
it is highly doubtful whether the film has adequate strength to resist 
the forces involved. 

However, if the film is propelled continuously in a straight line and 
at a constant speed of acceleration, considerably higher velocities can 
be obtained before the tensile strength of the film base becomes a lim- 
iting factor. In order to be able to project this film in a standard pro- 
jector there must be some control exercised to hold the film in the 
image plane and to hold the picture frame in synchronism with the 
perforations. The latter condition is easily obtained by the use of a 
sprocket in contact with the film and located so as to introduce the 
least amount of strain on the continuously moving film. 

An aperture plate or shoe could be used to hold the film in the image 
plane, but it is difficult to avoid the danger of too much friction in an 

Sept., 1945 



effort to guide the film at these high speeds and still prevent weaving 
and "breathing" at the image plane. 

The film control in the Fastax cameras is very simple and yet highly 
efficient. The film path has been arranged so as to permit the 
sprocket holes to engage at least 10 pairs of sprocket teeth at all times 
as the film travels around the sprocket from A to C, Fig. 1. This 
secures the film firmly on the periphery of the sprocket as it passes 
the aperture at B, minimizing the effect of the centrifugal forces that 
would tend to throw the film out of the image plane and cause the 
pictures to go in and out of focus, or "breathe." 

The sprocket is coupled directly to the driving motor and is used to 
propel the film through the camera by its 2 rows of specially designed 




/ ^ 


FIG. 2. 

and accurately cut teeth. The lateral weave of the film is practically 
eliminated by having the width of one row of teeth (A) on the sprocket 
almost the width of the film perforations, Fig. 2. These features 
assure uniform travel of the film at the high speeds of operation in- 
volved, but the means used make it necessary for the camera to em- 
ploy accurately perforated film that is within 0.5 per cent of full pitch. 

In addition to smooth film movement at high speed, it is necessary 
to provide suitable optical compensation to allow the image to follow 
the movement of the film during the exposure period for each frame. 

There are several methods of optical compensation that have been 
used by designers in the past. One method is the use of a series of 
matched lenses 2 - 3 - 4 of high aperture located around the periphery of 
a large wheel which rotates in front of the film. The lens wheel is 
driven at a rate of speed equal to the film as it passes the picture 
aperture. In this manner, one exposure is made through each lens as 

174 H. J. SMITH Vol 45, No. 3 

it moves downward at the same speed as the film; causing the image 
to remain stationary with relation to the film. If the lenses travel 
in an arc, there will be some transverse displacement of the image; 
however, there have been several methods devised to overcome this. 5 
There still remain difficult and critical adjustments of each lens and 
also the problem of securing uniformly matched lenses in quantities. 
Another method of optical compensation provides for the down- 
ward motion of the image by reflections from mirrors. 6 ' 7> 8 A re- 
volving drum fitted with mirrors rotates behind the lens. As each 
mirror face comes into position the image from the lens is reflected to 
the film which is traveling downward at the same rate of speed as the 

FIG. 3. 

reflected image, and one frame is exposed from each mirror face. The 
adjustment of these mirrors is very critical and the quality of the re- 
flecting medium and the accuracy of the surfaces involved must be 
very carefully controlled. This method requires the use of long 
focus lenses, as the distance from the rear element of the lens to the 
mirrors and then to the film is considerable. 

A third and simpler method is the use of a rotating optical compen- 
sating prism interposed between the lens and the film plane. This 
method has been used successfully in several other high-speed motion 
picture cameras. 9 " 14 The use of rotating parallel plane glass blocks 
has been discussed in previous articles, 5 ' 15> 16 but these generally 
apply to the rotating prism for use in continuous projectors where 

Sept., 1945 



the speed of rotation is not high, and where large-size prisms having 
many faces are used. 

Increasing the number of prism faces reduces the amount of the 
angular tilt of each face as it rotates past the aperture, thereby im- 
proving the definition of the refracted image. Unfortunately, in- 
creasing the number of faces of the glass polygon and still maintaining 
the same vertical dimension of the faces calls for increased diameters. 
For example, a 4-sided prism having faces with the vertical dimen- 


2 8 88 S '3 '8 s 











3 6 9 \2 15 18 


FIG. 4. 



sion approximately 0.265 in. high and a diameter of 0.554 in. would 
have its diameter increased to 1.10 in. for 8 faces, and 1.65 in. for 12 
faces, etc. As the diameter of the glass block increases, it becomes 
more difficult to secure suitable glass having the proper optical quali- 
ties required for making the prism. The difficulties of maintaining 
accurately the angular and linear dimensions of the prism are in- 
creased. The prism becomes bulkier with the increased mass, and in 
turn increases the danger of rupture generated by the centrifugal 
forces involved at high speeds of rotation. 

The advantages obtained by the increased number of faces of the 
prism were utilized in the camera under discussion by designing a 



Vol 45, No. 3 

prism having 4 pairs of polished faces forming an octagon. The 
diameter was held to a minimum by reducing the vertical dimension 


FUG. 5. 

of the faces to approximately 0.125 in., which is suitable for 8-mm 
pictures. The angular dimensions are held to very close tolerances 
to assure uniform refraction of the image during rotation of the prism. 







10 20 30 40 50 60 



FIG. 6. 





The index of refraction, the location of the prism with respect to 
the sprocket, and the dimensions of the prism are computed to give 
the correct movement of the image as the film moves continuously 

Sept., 1945 



past the aperture. Fig. 3 shows a plane parallel glass plate repre- 
senting any 2 parallel faces of the prism. If a light ray, A, enters the 
plate at the angle of incidence 7, it passes through the plate at the 
angle of refraction R, and emerges parallel to the angle of incident 




FIG. 7. Typical arrangement of lighting units for high-speed photography. 

ray, but is displaced by the distance SS'. The deviation of SS' pro- 
duced by the thickness of the plate T is found by 


T sin (7 - R) 
cos R 

Fig. 4 shows a curve of the displacement SS' at different angles of 
incidence. This curve is linear up to 15 degrees. Therefore, to obtain 



Vol 45, No. 3 

uniform displacement of the image the tilting angle should not exceed 
this amount. 

FIG. 8. Frames enlarged from a strip of 8-mm film taken at 8000 pic- 
tures per sec at //8, showing the action of a telephone fuse known as the 

The 8-sided prism rotates through 45 degrees for each picture frame 
on the sprocket. A point on the film moves 0.0375 in. past the aper- 
ture during this portion of a rotation. An angle of tilt of 1 l l / 2 degrees 

Sept., 1945 8000 PICTURES PER SECOND 179 

will give an image displacement of this amount. The angle of tilt is 
controlled by a housing which surrounds the prism. This housing 
has 8 apertures centered on the prism faces. Each aperture cuts off 
the incident rays at the selected angle of the outgoing face and does 
not permit the rays to pass through the prism until the incoming 
prism face has reached a corresponding angle of tilt, as shown in 
Fig. 5. 

The prism housing in conjunction with a front and rear aperture 
plate acts as a barrel- type shutter. The exposure time is controlled 

FIG. 9. Fas tax camera with door removed 
showing compensating prism, sprocket, and film 
path. The open box around the compensating 
prism contains the front and rear aperture. 

by the speed of rotation of the prism housing. If the prism rotates at 
60,000 rpm, pictures are taken at the rate of 8000 pictures per sec, 
and each picture receives an exposure of approximately 33 micro- 
seconds. If the prism rotates at 7500 rpm, the taking rate will be 
1000 pictures per sec, and each picture will receive an exposure of 
approximately 350 microseconds. 

The interposition of the prism between the lens and the film re- 
quires the use of lenses having a back focus of sufficient length to per- 
mit focusing the image on the film. The shortest focal length lens 
that may be used with this camera is 35 mm. 

The motive power to drive the camera is supplied by two V^hp, 



Vol 45, No. 3 

series-wound, universal-type motors having high starting and acceler- 
ating torques, and designed for high-speed applications where the 
load is directly connected and the duty is intermittent. One motor 
drives the sprocket to which the compensating prism shaft is geared, 
and the second motor drives the take-up reel. This large amount of 
power is used so as to accelerate the film to its maximum speed in the 
shortest possible time. By this means, it is possible to obtain pictures 
at maximum taking speed on at least 60 per cent of the 100-ft rolls of 
film used in the camera. The acceleration characteristics of the 
camera as finally constructed are shown in Fig. 6. 

FIG. 10. View with door open shows the film guard for the take-up reel, and 
the prism for the reflex viewer. 

Pictures taken at a rate of 8000 per sec require the highest order of 
illumination. The most convenient type of lighting units for this 
purpose are the tungsten filament "sunspots" found in any well- 
equipped motion picture studio. The intensity of these lights is in- 
creased for high-speed work by operating the lamps as close to the 
melting point of tungsten as is practicable. The lamps are often re- 
located with respect to the focus of the reflector, so as to focus at 
shorter distances, thereby permitting the focusing of the filament 
directly on the subject. The illumination required varies according 
to the area to be photographed, the color of the subject, and the tak- 

Sept., 1945 8000 PICTURES PER SECOND 181 

ing speed of the camera. In general, other factors being constant, 
the amount of light required will be in direct proportion to the taking 
speed, thus approximately 500 times as much light will be required to 
take pictures at a rate of 8000 per sec, as at 16 per sec. At the slower 
taking speeds, the intensity of light need not be so high; thus, with 
the camera operating at 2000 pictures per sec, photofloods such as the 
R2- and the R3-type lamps can be used. Pictures have been made at 
speeds in this range using a sequence of photoflash bulbs focused on 
the object and fired in rapid succession to keep the intensity peak 
constant. Arc lights may also be used in some cases. 

FIG. 11. Fastax camera closed. The view 
finder is mounted on the door. 

The focused lights are arranged to accent the proper modeling of 
the subject, so as to be sure that the moving parts under study are re- 
ceiving the full amount of illumination, and that the shadows of ad- 
jacent details are not obscuring the field of view. A typical arrange- 
ment of lighting units is shown in Fig. 7. 

When the proper technique is used in lighting the subjects, the 
Fastax camera will produce clear, steady pictures that can be pro- 
jected on any standard 8-mm projector, and have a fair depth of 
focus and good definition. The type of pictures obtained with this 
camera is illustrated in Fig. 8. These pictures were selected from a 
strip of 36 consecutive frames enlarged from an 8-mm high-speed 
motion picture film taken at 8000 frames per sec with the lens 

182 H. J. SMITH Vol 45, No. 3 

stopped down to//8, and show 0.0045 sec of action of a telephone- 
type fuse known as the "grasshopper." 

The final design of the camera is illustrated in Figs. 9, 10, 11, and 12. 
It is approximately 12X10X10in. high and weighs about 40 lb, and 
is suitable for operation either in the laboratory or in the field. 


FIG. 12. Motor side of Fastax camera. The 
motor on the left drives the take-up spindle, and the 
one on the right drives the sprocket and compensating 

A substantial number of Fastax cameras have been made and are 
now in active use by industrial organizations, research laboratories, 
and various branches of the Armed Services. 


1 EYLES, E. D.: /. Sclent. Instruments, 18 (Sept., 1941), p. 175. 

2 CONNELL, W. H.: /. Scient. Instruments, 4 (1926), p. 82. 

3 JENKINS, C. F. : "Motion Picture Camera Taking 3200 Pictures per Second," 
Trans. Soc. Mot. Pict. Eng., 17 (Oct., 1923), p. 78. 

4 JENKINS, C. F.: "The Jenkins Chronoteine Camera for High-Speed Motion 
Studies," Trans. Soc. Mot. Pict. Eng., 25 (Sept., 1926), p. 25. 

6 TUTTLE, F. E., AND REID, C. D.: "The Problem of Motion Picture Projec- 
tion from Continuously Moving Film," /. Soc. Mot. Pict. Eng., XX, 1 (Jan., 1933), 
p. 3. 

6 "The Zeiss-Ikon High-Frequency Camera," Brit. J. Phot. (Sept. 27, 1935), 
p. 617. 

Sept., 1945 8000 PICTURES PER SECOND 183 

7 SUHARA, T.: Proc. Imperial Academy Tokyo, 5 (1929), p. 334. 

8 SUHARA, T., SATO, N., AND KAMEI, S.: "Report No. 60," Tokyo University 
Aeronaut. Res. Inst. (May, 1930), p. 187. 

9 TUTTLE, F. E.: "A Non-Intermittent High-Speed 16-Mm Camera," J. Soc. 
Mot. Pict. Eng., XXI, 6 (Dec., 1933), p. 474. 

10 HERRIOTT, W.: "High-Speed Motion Picture Photography Applied to De- 
sign of Telephone Apparatus," /. Soc. Mot. Pict. Eng., XXX, 1 (Jan., 1938), p. 30. 

11 TOWNSEND, J. R.: Elec. Eng., 59 (Nov., 1940), p. 448. 

12 SMITH, H. J.: Bell Lab. Rec. XXH, 1 (Sept., 1943), p. 1. 

13 GALE, A. L.: Movie Makers (Feb., 1944), p. 54. 

14 JONES, L. A.: Phot. J., 84 (Apr., 1944), p. 118. 

TAYLOR, H. D.: Proc. Phys. Soc., 49, 6 (1937), p. 663. 

16 EHRENHAFT, F., AND BACK, F. G.: "A Non-Intermittent Motion Picture 
Project," /. Soc. Mot. Pict. Eng., XXXIV, 2 (Feb., 1940), p. 223. 



[Ed. Note. The following article by Dr. W. L. Everitt, from the Proceedings of 
the Institute of Radio Engineers, is being reprinted at the suggestion of a number of our 
members as it presents very clearly and completely the methods which should be used in 
presenting technical information, either to an audience or in a technical publication. 

We know that this excellent article can be of great assistance to our members and 
other contributors to the Journal of the Society of Motion Picture Engineers. \ 

Summary. The professional man has an obligation to give wide dissemination 
to new discoveries and developments. This may be done by publication and by oral 
presentation before technical groups. The proper presentation of a paper requires the 
co-ordination of four groups or individuals. A check list of their duties is given which 
may be used as a reference in the planning of technical sessions. 

The natural result of professional activities, particularly in science 
and engineering, is the development of new ideas and methods which 
are important, not only to the solution of a special problem but also 
to the general development of the art. It is not only the privilege 
but also the obligation of the professional man to make this knowledge 
available to other workers in his field and to the general public at the 
earliest possible time compatible with his own interest or that of his 
employer. Except when military security dictates otherwise, this 
information should be given as wide and early dissemination as pos- 
sible, since its originator cannot know to whom and to what extent 
this new knowledge may be useful for the benefit of mankind. 

It is to the advantage of the engineer and of his profession first to 
present publicly important developments and ideas of general utility 
in his field before a recognized professional society, where they may 
be discussed, developed, and perhaps questioned by his associates. 
By so doing, the engineer may be spared embarrassment or even dis- 
credit to himself or to his professional associates which sometimes fol- 
lows premature disclosure to the public press. 

* Reprinted from Proceedings of the I.R.E., 33, 7 (July, 1945), p. 423. 
** President, The Institute of Radio Engineers, Inc. 



The general dissemination of new technical ideas to the profession 
should be done in one or both of 2 ways : 

(1) By the presentation of a paper before a technical session of a 
professional society, 

(2) By the publication of a paper in the journal of a professional 
society or in a technical magazine. 

It is preferable that both methods be used for many developments. 
If the presentation is before a technical session where the attendance 
necessarily is limited, it is desirable to submit a paper for publication 
consideration to the society which sponsored the session, so that all 
members may be informed. When both methods are used, the pub- 
lished paper should include material which is developed in the discus- 
sion at the technical session, including a correction of errors which 
the discussion may bring forth. 

Papers which provoke oral discussion are the type which will give 
the most profit to the author, the society, and the public by presenta- 
tion first before a technical session. The papers committee in charge 
of the session, together with the author, should examine proposed 
papers with this point in view. The author should also keep this in 
mind in preparing his presentation. Unless early release is essential 
there is little profit to anyone merely in reading a paper before a 
group. Papers which require extended mathematical development, 
or detailed study to grasp their implications, are not suitable for pres- 
entation at technical sessions unless they can be briefed, and the 
important conclusions presented, so that the audience can participate 
in the discussion. 

A strong distinction should be drawn between reading and pre- 
senting a paper before a technical session. The reading, word for 
word, of material which later is to be published, without adequate 
opportunity for discussion, as has been done all too often, is a waste of 
everyone's time and fulfills no useful purpose. 

The proper presentation of a paper before a technical session re- 
quires the team action of at least 4 individuals or groups. They are 

(1) The Papers Committee, 

(2) The Arrangements Committee, 

(3) The Chairman of the Technical Session, 
(4} The Author. 

186 W. L. EVERITT Vol 45, No. 3 

A mutual knowledge of the duties and responsibilities of each mem- 
ber is desirable to obtain team action. 

The following suggestions are proposed for the duties of the team 
members in the conduct of technical sessions. While they are in- 
tended primarily for convention sessions, where there are a number of 
authors, most of the suggestions are applicable to section meetings. 
If you are a member of such a team, use this as a check list to be sure 
you are performing your functions. 

Duties of the Papers Committee. The Papers Committee, or an 
individual performing its functions, determines : 

(Z) The general theme of the technical session (industrial elec- 
tronics, radio antennas, etc.), 

(2) The time and date of the session, 

(3) The length of the session, 

(4) The chairman of the session (and contacts and secures him), 

(5) The number of papers, 

(6) The time allotted to each paper, 

(7) The proposed topics (examining them to be sure either that 
they can stimulate discussion, or that they should be pre- 
sented to advance the date of release of the information). 

The Papers Committee also 

(8) Contacts the author to obtain the paper, 
(Note : Items 7 and 8 may occur in either sequence.) 

(9) Handles all correspondence with the author, forwards sug- 
gestions as to the author's duties, determines from him what 
aids such as lanterns, blackboards, motion picture projec- 
tors, power outlets, manual assistance, and the like, he will 
require in his presentation. 

(10) Notifies the Arrangements Committee of the date and time, 
probable space required, and the requirements of the author 
in the way of lanterns, blackboards, and other supplies. 
Also indicates whether stenographic recording is desired. 

(11) Arranges for the introduction of the chairman, unless he is 
a section officer or otherwise acting in an official capacity so 
as to be known to the audience. 

Duties of the Arrangements Committee. The Arrangements 
Committee, or an individual performing its functions, arranges for : 

(1) The meeting place, 


(2} The mailing of notices and other publicity (unless under the 
jurisdiction of a separate committee), 

(3) Janitor service, 

(4) Projectors, blackboards, chalk, erasers, pointers, etc., 

(5) The lantern operator, 

(6) Smooth control of the darkening of the room from the lantern 
operator's position by remote control, or by signaling to an 
attendant at the room lighting switch, 

(7) Means for signaling between the speaker and the lantern 

(8) A public-address system where (and only where) needed. 
(When provided, the public-address system should have a 
lapel microphone if at all possible.) Check that it is in 
proper operating condition and that the gain is set properly. 

(9) Contact with the chairman of the session to apprise him of 
services provided and their operation, and to determine other 
services which may be desired. Inform him of any special 
conditions applying to the concluding of the session and va- 
cating of the premises. 

(10) Power outlets for demonstrations, 

(11) Assistance for bringing in and removing demonstration 

(12) Shipping instructions for the disposal of demonstration 
equipment after the meeting, 

(13) Chairs, 

(14) Tables, 

(15) Reading lights, 

(16) Distribution of material which is to be passed out to the 
audience (and its collection if necessary), 

(77) Recording of attendance, if desired, 

(18) Stenographic recording of discussion, if desired, 

(19) Meeting of the speaker at section meetings, if he comes from 
out of town, and making sure that his time is occupied to fit 
his convenience. (Have this done by a friend of the speaker 
if possible.) 

Duties of the Chairman of the Technical Session. The Chair- 

(1) Presides at the meeting and is responsible for the tempo and 
character of the whole session, 

188 W. L. EVERITT Vol 45, No. 3 

(2) Introduces speakers and outlines method of conducting dis- 

(3) Contacts authors, lantern, and light-control attendants in 
advance of the session to acquaint them with each other and 
with himself and to issue special instructions, including in- 
formation on 

(a) Facilities available, 

(b) Method of disposition of lantern slides, 

(c) Method of signaling operators, 

(d) Time schedule and method of adherence thereto, 

(e) Operation of public-address system, 

(/) Location of pointers, chalk, erasers, and any other sup- 
(g) Method of conducting discussion. 

(4) Ascertains from authors whether they know of members 
who will discuss papers, 

(5) Receives information from members who have prepared dis- 

(6) Conducts business where such is scheduled, 

(7) Is responsible for adherence to time schedule, 

(8) Calls for discussion, 

(9) Recognizes discussors in order, giving preference to those 
who have indicated preparation in advance. At conventions, 
he makes sure each speaker on the floor stands up and gives 
his name and business connection clearly. If he cannot be 
heard, the chairman should require him to come forward 
and face the audience and to use the public-address system 
if possible. Remember, a member of the audience in the 
center of the room has his back to half the audience. 

(10) Makes sure that questions from the audience are heard by 
all, and repeats them if necessary before the author replies, 

(11) Confines discussion to the topic, 

(12) Closes discussion when completed or at expiration of allotted 

(13) When desirable, requests that discussions, containing com- 
ments which are important, be submitted in written 
form to the author and to the Editor of the Proceedings. 
Forwards to the Editor discussions which are available in 
written form at the time of the meeting. 


Suggestions to the Author in the Presentation of a Paper before a 
Technical Session. It is beyond the scope of this article to develop 
the rules for effective public speaking, as there have been many 
publications on this subject. However, it is believed that adherence 
to certain technical and almost mechanical rules will improve the 
presentation of any paper. The following suggestions apply. 
Remember that a stimulating discussion is beneficial to the author, 
the audience, and the society. 

(1) Prepare a draft of the paper in advance. Keep in mind that 
your mission is to teach rather than to demonstrate how much 
you know. Be sure the paper is presented from a profes- 
sional viewpoint and not as an advertisement of your com- 
mercial connections. Make it interesting. Your efforts are 
wasted if the audience is bored. 

(2) Appear at the meeting sufficiently in advance of the scheduled 
time to meet and confer with the chairman, 

(3) Do not read the paper. Brief it and present it orally, empha- 
sizing its high points, especially any items over which there 
may be controversy. Be sure the material is in logical se- 

(4) Speak clearly and distinctly, look at your audience, and 
evidence your interest in them and in your subject. 

(5) Show the relation of the development discussed to the prog- 
ress of the art. 

(6) Distribute properly the time assigned to oral presentation, 
demonstration, and discussion. 

(7) Practice and time yourself beforehand. 

(8) Adhere to the time schedule. Expect the chairman to require 
you to do so. Use a watch which is constantly within your 
view. Modify your talk if you find you are running over- 

(9) Arrange for demonstrations if possible ; they always provoke 
more interest. 

(10) Notify Papers Committee of lanterns, blackboards, power 
outlets, and other facilities you will need: 

(11) Provide adequate illustrations, preferably on standard 
3y 4 - X 4-in. lantern slides. You may have to provide your 
own projector if nonstandard slides are used. Be sure the 
thumb marks are properly placed, and the slides are in order 
with the thumb marks facing the operator and in the upper 

190 W. L. EVERITT 

right-hand corner. Use a slide container which will keep all 
slides in order. 

(12) Use simplified block diagrams on slides where possible. Do 
not put too much detail on any slide; it is confusing in the 
short time during which it is shown. (This cannot apply to 
photographs of equipment.) 

(13) Make proper use of the public-address system. Keep at a 
constant distance from the microphone and in the same rela- 
tive position at all times. Even if you have a powerful voice, 
it is disconcerting when the public-address system fades out 
due to increased separation of the speaker from the micro- 
phone, or due to the turning of the head. 

(14) Never turn your head away from the audience while you are 
talking, even to point out items on the projection screen, 
unless you are wearing a lapel microphone. If necessary, 
point to the screen and then turn and speak into the micro- 
phone or towards the audience. 

(15) Use the facilities provided to signal the lantern operator and 
point to the screen. 

(16) Give credit to others who serve as a source for your state- 

(17) Make clear which statements you consider facts and which 
you consider conclusions or opinions. 

(18) In advance of the meeting, send copies of your manuscript to 
competent members who may attend the meeting. Ask 
them to come prepared to take part in the discussion. Do 
this particularly to individuals who may differ with you or 
oppose your views. Opposition may develop interest in 
your ideas and promote their acceptance if they are worth 
while, or save you from embarrassment later if you are in 
error. Notify the chairman of the meeting of any discussion 
you expect, and help him contact their source. 

(19) Rewrite your paper for publication, giving consideration to 
the points which were developed in the discussion. Send at 
least 3 copies to the Editor of the Proceedings as soon as 

(20) Secure copies of the discussions in typewritten form, if pos- 
sible. The chairman may assist you in this. 

Again, and above all, speak clearly and logically and adhere to the 
time schedule. 



Summary. The very small fraction of the total lamp output which finally reaches 
the screen from a projector might lead one to question the design of optical systems of 
the type now in general use. Nevertheless there does not seem, with present light 
sources, much chance of radical improvement. For practical purposes efficiency, in 
the sense of the ratio of screen lumens to total lamp lumens, is of much less conse- 
quence than screen brightness and picture quality. The benefit from improved 
efficiency is not necessarily a brighter picture, but more often the ability to work with 
a lamp of lower wattage. A designer may be justified in increasing the light wasted 
within his projector, for the sake of obtaining a proportionately smaller improvement 
in screen brightness. 

Other writers have discussed design from the standpoint of producing the best pic- 
ture. The present paper is largely confined to the subject of calculating efficiencies 
for such interest as it may have in showing where the light losses occur. The conclu- 
sion is reached that an optical system, designed for maximum screen lumens, is doing 
well from the efficiency standpoint if it projects to the screen more than 2 or 3 per cent 
of the total lamp output. 

Approach to Problem. During the discussion of John A. Maurer's 
paper on "The Optics of Motion Picture Projection" at the SMPE 
Technical Conference in New York in May, 1943, it was brought 
out that only about 5 per cent of the light emitted by the source 
reached the screen. This is a surprisingly low figure and would on 
first thought lead one to question whether an optical system which 
did not do better than this had been well designed. Before reaching 
any inference of this nature, one should recognize that designers of 
optical systems are far more concerned about the total light reaching 
the screen than about the efficiency of the system, or ratio of screen 
light to total lamp output. If, for example, the designer can add 20 
per cent to screen brightness by adding 100 per cent to the lamp wat- 
tage, he will in general do so. He will have a better projector, al- 
though his system will be of lower efficiency, using the word efficiency 
in the strict sense of ratio of useful output to total supply. Stimulated 
by the discussion just mentioned, the writer has made some calcula- 

* RCA Victor Division, Radio Corporation of America, Indianapolis, Ind. 


192 E. W. KELLOGG Vol 45, No. 3 

tions in an effort to see where the 95 per cent goes, and believing that 
the calculations are of some interest, they are being offered in this 
paper. Briefly speaking, the calculations are to answer the questions : 

(1) What fraction of the total lumens radiated by the lamp would enter con- 
densing lenses of several different numerical apertures (or subtending different 
angles from the source) ? 

(2) What proportion of the light passing through the condenser lens goes 
through the picture gate, and what fraction strikes the surrounding surfaces? 

(5) Of the light which passes through the picture aperture, what fraction enters 
the projection lens? 

Among the references ' ~ 6 listed at the end of this paper the reader 
will find discussions of the general problems of projection optics and 
studies of various special problems. 

, . . . 



Cosine LO.W 
U = Region of uniform intensity. 
G = Region of graded intensity. 

FIG. 1. Angle subtended by condenser, and character of beam. 

Two Types of Optical Systems. There are 2 fundamental types 
of optical systems, in which, 

(1) The light source (filament or arc crater) is imaged by the condenser system 
in the picture gate; 

(2) The light source is imaged in the projection lens and the condensing lens 
system is preferably as close to the picture aperture as practical considerations 

It is also possible to design a system with the light source image be- 
tween the picture gate and the projection lens, and such systems are 
common, but it is thought to be unnecessary to attempt here to calcu- 
late such a system. The figures for light losses would in general be 

Collection of Light by the Condenser Lens. The first problem, 


which applies to both types of optical systems, is to collect as much 
of the light from the lamp as practically possible and concentrate this 
in the filament image just mentioned. This means designing the 
condensing lens to subtend a large angle from the lamp. Calculations 
were made for only 2 types of radiation characteristics, nondirectional 
(or uniform in all directions) and bidirectional, or obeying Lambert's 
cosine law, such as would result if the source were a thin flat plate of 
hot tungsten. The monoplane lamp is an approximation to this. 

Table 1 gives the fraction of the total radiation of the lamp which 
would be intercepted by condenser lenses working at //1. 5, //I, 
//0.85, and //0.75. The term /-number is here used as the ratio of 
diameter to working distance, or h/m as illustrated in Fig. 1. It has 
been assumed in the calculations that the light radiated in the oppo- 
site direction is not utilized. This is very nearly true of the biplane 
lamp, in which the principal contribution of the mirror is in filling in 
some low spots; but in the case of multifilament monoplane lamps, 
properly adjusted mirrors can add to the useful light by an amount 
which is probably of the order of 60-70 per cent. No allowance is 
made here for the imperfections of the lenses which would cause some 
further losses. 


Fraction of Total Light Intercepted by Condenser (Cols. 3 and 4} 


Included Angle (Degrees) Lamp Characteristics 

/ number Measured from Axis Nondirectional Cosine Law 

1.5 36.8 0.025 0.05 

1.0 53.0 0.052 0.10 

0.85 61.0 0.069 0.129 

0.75 67.4 0.084 0.154 

Requirements for Full Illumination. It has been assumed for 
the purpose of calculation that the projection lens must be utilized 
at its full aperture and also that all parts of the picture aperture 
must receive full illumination. In practice a compromise is usually 
made in the way of reduced brightness at the corners of the screen. 
This permits considerable reduction in the total amount of light re- 
quired, but the formulas I have used can be applied to such a case by 
simply assuming a reduced diameter for the picture aperture, this 
reduced diameter, being that of the portion of the picture which does 
receive full illumination. The calculation of light losses at the gate 
or projection lens is purely a geometric one involving 3 apertures, (a) 



Vol 45, No. 3 

the condenser lens (its exit pupil), (b) the picture aperture, and (c) the 
projection lens (entrance pupil). The condition of complete illumina- 
tion of the picture aperture and complete utilization of the projection 
lens imposes a size requirement on the condenser. The condenser 
lens must be large enough so that there will be no point within the 



51 = Lens stop. 

5 2 = Virtual image of Si. 

FIG. 2. Optical system, with filament imaged in projection lens. 

projection lens stop, from which the edge of the condenser lens pupil 
can be seen through the picture aperture. This cannot be true unless 
the rim of the condenser falls outside the diagonal lines d, d, shown in 
Fig. 3. Or, to express it differently, from any point within lens C of 
Fig. 3 the picture aperture should appear to be completely filled with 
light coming from the condenser lens. (See dotted lines through 
point P in Fig. 4.) If the conditions just described are met, the screen 

FIG. 3, Simplified version of Fig. 2. 

brightness will be the same and therefore the useful light will be the 
same, whether the filament is imaged in the projection lens or in the 
picture aperture or, for that matter, in between. In other words, it 
is possible to achieve the same final result with either system No. 1 or 
No. 2 (assuming the light source to be of such uniformity that No. 1 
can be considered). This is not saying, however, that the result will 
be obtained with the same sized lamp. 


To provide a specified screen brightness, the designer first chooses 
a projection lens of the desired focal length and of the necessary aper- 
ture. The condenser is then designed of sufficient diameter in rela- 
tion to its spacing from the picture aperture to comply with the re- 
quirements just discussed, and of as short focal distance on the lamp 
side as is considered practical, and with the filament magnification 
ratio thus determined, a lamp must be chosen constituting a suffi- 
ciently large source so that its image will fill the picture aperture or the 
projection lens (whichever the chosen system calls for). Spherical 

wasted I ight 
Filament imaged at b 

Filament imaged atrc. 
FIG. 4. Conditions for complete illumination of picture aperture. 

aberration in the condensing system may make it necessary to pro- 
vide some margin in condenser size, or else cause some loss of bright- 
ness, usually at the corners of the picture. The calculations pre- 
sented here do not make any allowance for spherical aberration, but 
this should scarcely alter conclusions as to the relative efficiencies of 
various arrangements. Moreover, we set out to ascertain what wast- 
ages are practically unavoidable, even with the best of lens designs. 

It is evident that if the design procedure just outlined is followed, 
the penalty for unnecessary wasting of light in the optical arrange- 
ments is the necessity of using a larger lamp to produce a given picture 

196 E. W. KELLOGG Vol 45, No. 3 

Calculation of Condenser Diameter. The first step in estimating 
light losses at the picture and projection lens apertures, is to calcu- 
late, for any assumed sizes and spacings of these apertures, how large 
a condenser lens will be required. The size of the condensing lens 
with the size and distance of the image of the filament produced by 
it determine the total light which the condenser lens must collect. If, 

a stands for the diameter of the condenser lens (exit pupil), 
b stands for the diameter of the picture aperture (diagonal if aperture is rec- 
tangular) , 

c stands for the diameter of the projection lens (entrance pupil) , 
x stands for the distance from condenser to picture aperture, 
y stands for the distance from picture to projection lens, 

the diameter of the condenser to meet the illumination requirements 
stated above is given by the following formula which is simply an 
algebraic statement of the geometrical construction shown in Fig. 3 : 

a = c - + b(- + lY 

y \y ) 

It is worthy of note that the required condenser size (for given values 
of b, c, x, and y) is independent of whether the filament is imaged in 
the gate or the projection lens. However, the total amount of light 
which the condenser must collect will not necessarily be the same in 
the 2 cases, the size of the lamp and distance from lamp to condenser 
being chosen to produce a filament image of the required size to fill 
the gate or the lens as the case may be. 

Light Losses at Picture and Projection Lens Apertures. Of the 
light collected by the condenser, only a fraction passes through the 
picture aperture and projection lens. It is next of interest to estimate 
how much is thrown away, either around the picture aperture or 
around the projection lens. There is a well-known expression for 
total light transmitted through 2 coaxial apertures. If a source has 
a brightness B candles per sq cm and area AI sq cm, it will radiate 
BA i lumens into each unit of solid angle (steradian) . If it illuminates 
an area A 2 sq cm / centimeters away, the solid angle subtended by 
A 2 is A 2 /l 2 . Then the number of lumens passing from AI through A 2 
is approximately (assuming/ 2 is substantially larger than 'A i or A 2 ) : 

L = B (1) 

This expression is applicable only when the optical arrangements 


are such that the light intensity is uniform at A\ and A^.* Such a 
condition holds when A\ is taken at a condensing lens and A 2 at the 
image of the source, provided the source is of substantially uniform 
brightness. If the source comprises a series of parallel filaments, it 
will not be uniform, but an average brightness may be taken for the 
source provided A 2 is large enough to include several coil images. 

The purpose of our calculations is to show what light losses are un- 
avoidable in the system and not to include any which are unnecessary. 
Therefore, it is assumed that the filament image is of the correct shape 
and just large enough to fill the picture aperture in the case of system 
No. 1, or to just fill the projection lens in system No. 2. The correc- 
tion for shape factor is discussed later, but without such a correction 
it is possible to draw certain general conclusions, for which purpose 
we shall for the present assume circular apertures throughout and 
assume a circular light source image. 

The total light transmitted by the condenser may be calculated, 
using formula (1), with A\ standing for the condenser lens exit pupil, 
A 2 the filament image area, and / the distance between them. 

The useful light is calculated from the same equation, letting A\, 
A 2 , and / stand for the picture gate area, the projection lens aperture 
area, and / the distance between them, or using the symbols shown on 
Fig. 3, the useful light is 

(0.78546 2 )(0.7854c 2 ) 


for a circular aperture at b. 

The ratio of total light transmitted by the condenser to the useful 
light, may be calculated for the case of a circular picture aperture and 
light source by the following formulas : 

Optical System Type 1 

Filament imaged in picture gate, just covering latter. 


* So that it may be said that there are rays from all parts of A i to every part of 
A%. If the intensity of the light beam is uniform throughout its cross section at A\ 
and at A 2 , then at positions between AI and A 2 and also for a short distance beyond 
A 2 , the light beam cross section has a central area of uniform brightness, as indi- 
cated in Fig. 1, and an outer zone in which the intensity tapers off to zero at the 
outer edge. Hence, the cross section of the beam at such a point may not be used 
in calculating transmitted light. When large angles are involved, or in other 
words unless / is several times V ' A\ or -\/At, formula (1) is approximate only. 

198 E. W. KELLOGG Vol 45, No. 3 

Optical System Type 2 
Filament imaged in projection lens, just filling latter. 


Useful L b \x + y 

The formulas serve well to bring out the conditions under which each 
system operates to best advantage. For minimum light loss, the 
second term within the parentheses must be made small. Thus, in 
system No. 1 the wasted light is minimized when the projection lens 
is large and the picture aperture small. There is a further advantage 
in working with the condenser well back from the picture gate. 

In the case of system No. 2, it is possible (with circular picture 
aperture) to reduce the wasted light to a small value by making x 
nearly zero or, in other words, placing the condenser adjacent to the 
picture aperture. There is objection to doing this because any dirt 
or scratches on the condenser would be sharply focused on the screen. 
Therefore, this term cannot be made equal to zero, and the ratio of 
projection lens to picture aperture diameter is an important factor, 
efficiency being favored by large picture aperture and small projec- 
tion lens, which is not the usual condition with motion picture projec- 
tors. However, with practical dimensions in 16-mm projectors, the 
balance is not seriously unfavorable to system No. 2, provided the 
condenser does not have to be too far back from the picture gate.* 
With incandescent lamps, the source is too irregular to image in the 
gate, and therefore we are practically limited to system No. 2. 

In system No. 1, the excess light goes through the picture aperture 
and is thrown away around the outside of the projection lens. In 
system No. 2, the wasted light is intercepted by the picture aperture 

There are further losses owing to the fact that the picture aperture 
is a rectangle instead of a circle and the fact that the image of the 
source may not be the same shape as the aperture in which the image 
is formed. System No. 1 is widely used with arc lamps, and the arc 
image must much more than cover the rectangular picture aperture 
in order to insure uniform illumination throughout the picture. In 
the case of system No. 2 there would be minimum loss with a circular 
light source whose image would just fill the projection lens, and so far 
as the sources are not circular, the source image must be oversize. 

* Compare, for example, the total to useful light ratio of Cases 4 and 5 with 
Cases 1 and 2. 


There is no way in system No. 2 to avoid the loss because of the fact 
that the pictures are rectangular. The total light required is the 
same as for a circular aperture whose diameter is equal to the diagonal 
of the rectangle. The useful light is reduced in the ratio of the area 
of the rectangle to that of the circle. 

Table 2 shows values of the ratio of total light transmitted through 
the condenser to that which is transmitted through the projection 
lens to the screen, calculated by formulas (2) and (3). No allowance 
is made for reflection losses at glass surfaces. 

The arrangement described by F. E. Carlson 4 is an interesting 
combination of the 2 systems, in which a cylindrical or toroidal surface 
is used in the condensing system to bring the filament image into the 
plane of the gate in the vertical direction (in which the uniformity of 
the source is excellent) while in the horizontal plane the filament is 
brought to focus in or near the projection lens. Mr. Carlson found 
a 15 per cent increase in screen brightness obtainable by this expedi- 
ent. Since it is presumable that the projection lens was filled in both 
directions before the introduction of the cylindrical surface, the extra 
light output may have resulted from the fact that the new system de- 
rived its light from the middle portions of the filaments which were 
hotter than the ends. 

Figs. 5 to 10 show the relative dimensions of the several systems 
compared in the table, and also the over-all efficiency of each arrange- 
ment, assuming the picture aperture to be 0.3 X 0.4 in. and the light 
source approximately circular, with cosine law distribution, and based 
on a condenser system working at //I with respect to the lamp, or 
collecting 10 per cent of the total lamp output. It is not exactly fair 
to assume the same collection angle for the condenser, for in system 
No. 1 the filament image must be formed nearer the condensing lens 
than in system No. 2, making it in general necessary to increase 
slightly the spacing between the filament and condenser, or else use a 
stronger condenser of the same diameter. 

Case 7 illustrates improved efficiency at the sacrifice of screen 
brightness, by showing the reduced ratio of total to useful light, which 
in system No. 2 is obtainable when the projection lens is slow. Ob- 
serve that for x = y and the projection lens twice the diameter of the 
maximum dimension of the picture aperture (Cases 1 and 4), system 
No. 1 shows no better efficiency than No. 2, but that with faster pro- 
jection lenses or with x exceeding y, system No. 1 increases in efficiency 
while system No. 2 loses. 



Vol 45, No. 3 

y 10 co 

O 00 b- OS O 






iO 1> CO 1> 

d r}H 00 CO 



CM o CM 



iO iO iO 



iO CM 

iO O iO iO CO 

d d d d d 

IO *O 


3 6 

5 8 

H I* 


O 4-> 



TO 4J 

G d 

'5 rt 



cu ^ 3 




If* S 

<L) ^ 

3 G T-H 

g.js I 

<* o > 

s >- 

a *g ZH 
^ .^ s 
"" -S *S 

2 ^ 


8 3 

_. o 

CO ^_, 

3 *g 
S o 

en - rj 

I 2 


en ^ 
"3 1> 


4^ ^ 


S ft 

4-> 00 

1 = 

d cu 

d o 

s s 



.S 3 
4 S 

d D 
X -^ 

n > S 
o S 

<u <u 

"2 '3 

"rj Q 1 


co 2 





w O 

~ en 

O D 


,x o 


d 5 

*a^ I 

OJ * J en 

O- C8 

.52 o 

43* 1> 

.*3 ^ 





Q OOOn-lr-l^H-lr-l R U ^ .?. ^ 






(L> S *C 


* -3 S 
"S 5 


3 4^ C 

o *-> <" 

rH W 




cJ ^s c^3 

O .S - 

: 2 13 

T=! =3 Q 



Even with the optimistic conditions assumed for these calculations, 
the over-all efficiencies are seen to be very low, so that the 5 per cent 
mentioned at the beginning of the paper would be excellent. Prac- 
tically obtainable efficiencies would be less than calculated because of 
glass-air surface reflections, condenser aberrations, and filament shape 

FIG. 5: 
Case 1 
Case 4 

FIG. 6: 
Case 2 



Eff. (%) 




FIG. 7: 
Case 3 


FIG. 8: 
Case 5 


FIG. 9: 
Case 7 

FIG. 10: 
Case 8 



factor. On the other hand, in an incandescent lamp system a prop- 
erly adjusted reflector would give an improvement not considered in 
the calculations, and partially offset the extra losses just mentioned. 
The reflector not only raises efficiency but also improves screen bright- 
ness and uniformity. 

No attempt has been made here to calculate efficiencies of arc light 
systems, but considerably higher efficiencies are obtainable, partly 

202 E. W. KELLOGG 

because the source inherently radiates the major part of its light in 
one direction, and partly because the source is sufficiently uniform to 
be imaged at or near the picture aperture. It is seen from formula 
(2) that a relatively large projection lens and small picture aperture 
(as is true of 35-mm projectors) as well as large ratio of x to y are 
favorable to the minimizing of light losses in system No. 1. 


1 HARDY, A. C.: "The Optics of Motion Picture Projectors," /. Soc. Mot. 
Pict. Eng., XIV, 3 (Mar., 1930), p. 309. 

2 MILI, G., AND COOK, A. A.: "Condensers for 16-Mm Optical Systems," /. 
Soc. Mot. Pict. Eng., XXVI, 6 (June, 1936), p. 603. 

3 MILI, G.: "Effect of Light-Source Size with 16-Mm Optical Systems," /. 
Soc. Mot. Pict. Eng., XXVIII, 2 (Feb., 1937), p. 164. 

4 CARLSON, F. E.: "A Higher Efficiency Condensing System for Picture Pro- 
jectors," J. Soc. Mot. Pict. Eng., XXXI, 2 (Aug., 1938), p. 187. 

5 CARLSON, F. E.: "Properties of Lamps and Optical Systems for Sound 
Reproduction," /. Soc. Mot. Pict. Eng., XXXIII, 7 (July, 1939), p. 80. 

6 COOK, A. A.: "Optics of Projectors for 16-Mm Film," /. Soc. Mot. Pict. 
Eng., XVIII, 4 (Apr., 1932), p. 461. 


T. Y. LO** 

Summary. To use motion pictures to enlighten international understanding, 
foreign-version films have to be considered in order to stimulate the local interest. 
Instructional and informational films can be changed into other languages either by 
narration or captions. It will be of more interest to the audience if life-dialogues in 
instructional films are synchronized with actors' lip movements into other languages, 
provided it is well done. Entertainment films are restricted to captions translating 
the dialogue. 

The author proposes that the picture area be reduced to provide a space at the 
lower margin of the screen for captions. A reduction printer must be developed for 
the process. 

General George C. Marshall recently said that the second World 
War has seen the development of 2 new weapons: the airplane and 
the motion picture. In the China Security Plan offered to the Allies 
at Dumbarton Oaks in Washington, the motion picture is again 
emphasized as a vital medium in attempting to promote international 
friendship and understanding for maintaining world peace. 

If the motion picture can be used effectively to kill and destroy, 
how much more can this medium do to teach men how to build, both 
physically and mentally, for a great future. The motion picture can 
show well the appearances of things and actions, but without the 
necessary interpretation to enable the audience to comprehend the 
motives and meanings underlying the actions, much of the picture 
will be meaningless. The value of the spoken word will be lost; in- 
stead of assisting to bring peoples together the very reverse may 
happen the audience attempting to understand and not being able 
to do so, may become dissatisfied and even disgusted, and may prob- 
ably give up all future attempts to understand. One may say, in 
effect, "Why should I be annoyed? I can't understand it, and the 
producer doesn't seem to care whether I do or not." 

* Presented Oct. 18, 1944, at the Technical Conference in New York. 
** Chinese Supply Commission, Washington, D. C. 


204 T. Y. Lo Vol 45, No. 3 

Another point of importance is the manner in which the interpret- 
ing is done. It is bad to splash an interpretive caption across a beauti - 
ful picture the photography is ruined and the attention is distracted. 
If the interpretations are on a separate screen alongside the picture, 
the eyes must dance a jig back and fprth from the picture screen to 
the caption screen; there is a great contrast of light and dark that is 
annoying and fatiguing, and altogether the whole thing is unsatis- 

Educational and informational films form the subject of this paper, 
not so much from the point of view of content or the manner of pre- 
senting the material in picture and word form, but from the point of 
view of making such films fulfill their complete purpose when pre- 
sented to a foreign audience. In adapting educational and informa- 
tional films for foreign audiences only the sounds are changed sound 
track, or titles, or both from the one language to the other. How- 
ever, it may not always be well to change the sound track; for ex- 
ample, if we were to show a Chinese audience a newsreel of Franklin 
D. Roosevelt making a speech in English, of course, because I doubt 
that Mr. Roosevelt could speak Chinese it would be unreal and even 
misleading to use a sound track in Chinese synchronized to Mr. 
Roosevelt's lip motions. The original sound track in English must 
be used in such an instance, even though the audience will not under- 
stand it. The only thing that can be done is to use superimposed cap- 
tions or translations of what is being said. 

There are 2 ways in which titles or translations are added to the 
films nowadays. One method has already been mentioned : the trans- 
lation is photographed on a separate film from the picture film, and 
the two are printed together onto a duplicate negative to be used for 
making foreign release prints. This method puts the captions or 
translations on the picture area, as discussed previously. Another 
way of accomplishing the same effect is used currently in China : zinc 
engravings of the captions are made, and are pressed against the 
film, frame by frame, in a step printer, with a special chemical solu- 
tion for removing the emulsion. This method has been used in cases 
where the important release prints were limited to only a few say, 
two or three in order to avoid large import duties and also to permit 
making the translations to suit the local uses and needs. 

Educational pictures in which the dialogue consists of technical in- 
struction are suitable for lip-movement synchronization. The U. S. 
Chinese Version Training Films produced by the U. S. Army Signal 


Corps in cooperation with the Chinese were lip-synchronized, and 
were well received in China. The Chinese Army, especially, found 
much interest in these films. The job of synchronizing lip movements 
is difficult, but the audience finds additional interest in the film when 
a lecturer, foreign to the audience, uses a language other than his own 
to give instruction provided the synchronization is well done. If it 
is not well done, it may as easily distract attention. 

FIG. 1. Foreign -version picture with Chinese caption. 

In order to synchronize lip movements it is necessary to divide the 
film into short sections the shorter the easier. Each section to be 
synchronized is made into a loop and is run over and over again while 
the synchronizers become familiar with it, and rehearse it over and 
over again until their lip movements synchronize perfectly with the 
lip movements of the actors on the screen. It has been found helpful 
also to draw curves in ink on the work print indicating the opening 
an'd closing of the lips. The synchronizer sees this curve moving right 
and left over the picture as he views the work print. This procedure 
helps the synchronizer to follow the tempo, and also eliminates the 
difficulty of seeing accurately the lip movement in long shots. 


T. Y. Lo 

Vol 45, No. 3 

As indicated before, superimposed titles and lip synchronization 
are not suitable for entertainment pictures. Entertainment pictures 
have dramatic values and special treatment for creating realism, and 
any disturbance of the senses by artificial sounds or vocal cadences 
will destroy the illusion that the producer hoped to instill in the minds 
of the audience. 

FIG. 2. Chinese picture with English caption. 

It is hardly necessary to raise again the question of how much of the 
photographic beauty of the picture is spoiled by putting translations 
into the picture area. And, in addition, it is often difficult to read 
such captions, when they are printed in white on a light background. 
There is also the difficulty of trying to synchronize the lip motions, 
as even a perfect synchronization cannot retain the dramatic quality 
and realistic feelings. Imagine trying to put into another language 
some of the emotional difficulties that Bette Davis portrayed in Mr. 
Skeffington! Some American pictures I have seen that were dubbed 
in French have been entirely ruined dramatically. It is my belief 
that entertainment pictures should neither be dubbed in foreign 



languages nor have foreign-language captions or titles printed in the 
picture area. Such things should be restricted to educational and in- 
formational films. 

With regard to entertainment films, Figs. 1 and 2 show an arrange- 
ment proposed by the author which will avoid most of the difficulties 
discussed above. It is the simple expedient of optically reducing the 
picture during printing so as to leave sufficient space beneath the 
picture for titles and translations. The picture is reduced in the same 


18.00 MM 

1.48 MM 

1.64 MM 

20. 96 MM 

FIG. 3. Reducing dimensions based upon SMPE recommended projector 
aperture dimensions. 

proportions as given in the SMPE Recommended Practice for pro- 
jector aperture. 

The reduction in size of the picture will be only 14 per cent that 
is, the reduced picture will be about 7 /s the size of the standard picture, 
and will have the same proportions as the original. With this reduc- 
tion a lower margin of 15 in. will be left at the bottom of the picture 
on the screen image 20' 2" X 14' 6", which is the size of image thrown 
by a 4V2-in. lens at a distance of 110 ft. Fifteen inches will be ade- 
quate for 2 lines of Chinese characters and 3 lines of English. The 
English letters can be made about 4 in. high (allowing separation be- 
tween lines) and 110 ft will subtend an arc of about 10 sec, which is 
more than sufficient for legibility. 


T. Y. Lo 

If the theater manager for some reason or other does not like the 14 
per cent reduction in size of picture image, he can install the next 
larger size screen and use a 4-in. lens whenever he shows the foreign- 
version films. The 4-in. lens will provide 11 per cent more magnifica- 
tion, so that the loss of screen image size will be only 3 per cent. The 
reduction in image brightness in either case will not be important. If 
the theater is to show both native and foreign- version films, a system 
of movable masks can be used around the edges of the screen. This 
may not be necessary, however, as it is proposed to print the captions 
and translations so as to appear on the screen in white on a black back- 
ground in harmony with the black masking. 

FIG. 4. A suggested method for printing foreign-version picture and caption. 

Based upon the reduction of 7 /s size (more exactly 86 per cent), the 
actual size of the picture image on the film will be 18.00 X 13.12 mm. 
The lower margin will be 1.64-mm wide and the top margin 0.48 mm. 
The side margins will be 1.48-mm wide. The total frame area will 
be the same as the SMPE Recommended Practice, viz., 20.96 X 
15.24 mm. Fig. 3 shows these dimensions. 

A new optical printer is needed for such a process, the design of 
which should be left to the printer experts. Several arrangements 
should immediately suggest themselves, such as the use of masks or 
by simultaneous projection of picture and titles by separate optical 
systems (see Fig. 4) . The new optical printer should find world-wide 
application, and all the countries of the world could produce their 
own foreign- version films locally from duplicate negative supplied 
by the producing companies. 





Summary. // is impossible to attain absolute dimensional stability of an image 
on ordinary photographic films. There will be slight, but nevertheless measurable, 
distortions of the image owing to the loss of volatile materials and to the expansion or 
contraction which accompanies changes in moisture content. Likewise, a small 
amount of image distortion occurs frequently in printing. 

These distortions are usually quite small and can often be neglected. In certain 
specialized fields, however, they cannot be overlooked. In one part of this paper the 
various sources of these image distortions are discussed individually and suggestions 
are offered as to the manner in which these distortions may be minimized. 

Another part of the paper is devoted to a discussion of the drying of motion picture 
film and the difficulties frequently encountered when the film is reeled with the emulsion 
layer in equilibrium and the base in partial equilibrium with the air in the drying 

In a previous paper, J. M. Calhoun 1 has discussed the physical 
properties of motion picture film and the extent to which these proper- 
ties are influenced by heat, moisture, and other factors. It is the 
purpose of this paper to consider the relation of a few of these film 
characteristics to certain problems in the laboratory and in the theater 
projection room. The problems selected for discussion are image 
distortion and film defects resulting from insufficient drying. 

As a rule, a small amount of distortion of either the negative image 
or the positive image does not detract noticeably from the usefulness 
of the motion picture print. There are, however, certain special 
fields in which the matter of image distortion has become increasingly 
important. These fields include color photography, special-effects 
photography, and certain applications of photography for military 

An image on photographic film should not be considered as stable 

* Presented Feb., 23, 1944, at a meeting of the Atlantic Coast Section of the 
Society in New York. 

** Eastman Kodak Company, Kodak Park, Rochester, N. Y. 




Vol 45, No. 3 





as one on glass or steel. Actually, the image on a piece of film is 
changing in size or shape continually throughout its life. Distortion 
of the photographic image is generally the result of shrinkage owing 
to loss of volatile materials, expansion or contraction produced by 
changes in moisture content, and failure of the printer to transfer the 
exact dimensions of the negative image onto the positive film. The 
magnitude of these distortions can be illustrated by selecting a test 

object of known dimensions, 
photographing it, and follow- 
ing the dimensional changes in 
the photographic image of the 
object throughout the life of 
the film. A suitable object 
may be a test target, as illus- 
trated in Fig. 1, constructed 
on a dimensionally stable ma- 
terial such as glass. The target 
is 1.378 in. (35 mm) wide and 
of any convenient height. It 
consists of 4 fine lines A, B t C, 
and D. For convenience, line 
A is located 0.2000 in. from 
edge 0, and line B is located 
1.0000 in. from edge 0. Lines 
C and D are 0.5000 in. apart. 
The image of this target is now 
transferred to a piece of 35-mm 

negative film by contact print- 
FIG. 1. Test target on glass for image . ,. ,. f . 

distortion study. m > perfect alignment of target 

and film at edge being main- 
tained when the exposure is made. There has now been produced 
on the 35-mm film a latent image whose dimensions are identical 
with those of the test target. The film is then processed. 

Dimensional Changes in the Negative Film. It will be found that 
the dimensions of the developed image on the film do not coincide 
exactly with those of the target. If care is taken to bring the nega- 
tive film to equilibrium with air at the same relative humidity after 
processing as that of the air with which it was in equilibrium at the 
time of exposure, any dimensional changes will be caused solely by 
processing. The permanent processing shrinkage of Eastman motion 



picture negative film (nitrate base) is of the order of 0.10 per cent in 
length and 0.15 per cent in width. (For comparable shrinkages of 
Eastman motion picture negative film, safety base, see "The Physical 
Properties and Dimensional Behavior of Motion Picture Film" Table 
4. 1 ) Therefore, upon measurement of the image lines after process- 
ing, it will be found that line A has moved 0.0003 in. toward edge 
from its original position, and that line B has moved 0.0015 in. toward 
edge 0. Lines C and D are now 0.4995 in. apart. 

Ordinarily, it is not necessary to maintain the processed negative 
in equilibrium with air at the 
same relative humidity as that 
with which it was in equilib- 
rium at the time of printing. 
However, in the special cases 
noted, the changes in dimen- 
sions which accompany changes 
in moisture content of the 
negative become an important 
factor. It may be assumed 
that most negative raw stock, 
when removed from the con- 
tainer, is in equilibrium with 
air at about 60 per cent rela- 
tive humidity. If the nega- 
tive, after processing, is stored 
in a dry place, there will be a 
further shrinkage of the tempo- 
rary or reversible type caused 

by loss of moisture. Thus, if the negative were stored in a vault 
in which the relative humidity of the air averaged 20 per cent, there 
would be a contraction of about 0.20 per cent lengthwise and 0.25 
per cent widthwise owing to the loss of moisture. These dimen- 
sional changes are in addition to the permanent shrinkage caused 
by development. Consequently, line A has now moved a total of 
0.0008 in. toward edge 0, line B has moved a total of 0.004 in. to- 
ward edge 0, and lines C and D are 0.4985 in. apart. 

This is not all. There will be further permanent shrinkage on 
storage, owing to a gradual loss of traces of solvent or plasticizer from 
the processed negative. The shrinkage of Eastman nitrate motion 
picture negative film resulting from loss of material upon storage 

FIG. 2. Film paths on a typical 
continuous Printer: A Path of the 
negative; B Path of the positive. 

212 R. H. TALBOT Vol 45, No. 3 

under normal conditions for one year is approximately 0.3 per cent 
lengthwise and 0.4 per cent widthwise. Consequently, after one 
year, under normal keeping conditions, the following shrinkages of 
the negative may have taken place : 

Lengthwise, Widthwise, 

per cent per cent 

Processing shrinkage 0.10 0.15 

Humidity change contraction . 20 . 25 

Keeping shrinkage (1 year) 0.30 0.40 

0.60 0.80 

The positions of lines A and B will be 0.0016 in. and 0.008 in., respec- 
tively, closer to edge than in their original positions, and lines C 
and D will be 0.4970 in. apart. 

Ordinarily, distortions of this magnitude have little or no signifi- 
cance when the negatives are used for normal printing. In special 
cases where distortions of this amount are objectionable, the dimen- 
sional changes accompanying variations in moisture content can be 
kept at a minimum by reconditioning the negative to 60 per cent 
relative humidity. The shrinkage during keeping can be minimized 
by storing the negatives in sealed containers at about 50-60 F and at 
40-50 per cent relative humidity, when not in use. 

Dimensional Changes in the Positive Film. The positive print, 
after exposure to the negative, is subject to the same general dimen- 
sional changes which have been described in the case of the negative. 
That is, there will be a shrinkage upon development, dimensional 
changes produced by variations in the moisture content, and shrink- 
age upon storage owing to the loss of solvent or plasticizer. It would 
be possible, therefore, in the very unusual case in which all dimensional 
changes in the negative at the time of printing were as great as possi- 
ble and all additive, and in which subsequent to printing the dimen- 
sional changes of the positive were as great as possible and all addi- 
tive, to have displacements of certain lines in the print of the order of 
0.02 in. from those of the original subject. 

Image Distortions Produced in Contact Printing. In contact 
printing, the negative and positive are brought together, emulsion 
layer to emulsion layer, and the exposure is made. At first thought, 
one would expect that an exact reproduction of the negative image 
would in all cases be recorded on the positive. In a step printer, in 
which each frame of the negative is held in contact with the positive 


film in a flat plane when the exposure is made, this is substantially 
true. In the case of a continuous printer, in which the exposure is 
made after the negative and positive are brought in contact on the 
curved surface of a drum, the image of the negative is not transferred 
unchanged in all respects to the positive. In Fig. 2, which illustrates 
a portion of a continuous printer, it may be seen that the negative 
film travels in a circular path which is shorter than that of the 
positive. Thus, for a drum of 1.887-in. radius, as on the standard 
Bell and Howell Model D printer, and assuming that the neutral axis 

FIG. 3. Compression and extension effect on bending 
an elastic material. 

of the film is at the center, the path of the negative is 0.317 per cent 
shorter than that of the positive. If the perforation pitch of the 
positive is equal to that of the negative, the positive film must gain 
its greater distance of travel by slipping past the negative film during 
exposure to the extent of 0.317 per cent. In practice, it is usually 
found that the positive film is unshrunk and that the negative, be- 
cause of processing and the time interval which exists between proc- 
essing and printing, has shrunk nearly 0.2-0.3 per cent. When 
such is the case, printing on a continuous printer is quite satisfactory. 
The pitch of the negative must be approximately 0.3 per cent less 

214 R. H. TALBOT Vol 45, No. 3 


than that of the positive, or the picture will appear slightly unsteady 
and the sound will be slightly distorted. 

These considerations in regard to continuous printing are quite 
obvious and certainly well understood. There is, however, another 
effect which takes place when the exposure is made while the negative 
and positive are in contact on a curved surface which is not so gener- 
ally appreciated. When any elastic material is bent, the concave 
side is compressed, and the convex side is stretched (Fig. 3). There- 
fore, on a continuous printer, the emulsion side of the positive film is 
compressed, and the emulsion side of the negative is stretched at the 
instant of exposure. Obviously, this effect will make it impossible 
to reproduce exactly longitudinal dimensions of the negative onto the 
positive by this manner of printing. Assuming again that the 
neutral axis of the film lies at its center, the positive emulsion surface 
has been compressed longitudinally 0.317/2 or 0.158 per cent, and 
the negative emulsion surface has been stretched 0.158 per cent at the 
instant of exposure. When the positive film returns to its normal or 
flat position, the emulsion surface will expand 0.158 per cent, giving a 
total longitudinal increase in dimensions of 0.317 per cent. This 
degree of distortion is so small as to be inconsequential in the case of 
ordinary usage. It is only in connection with the special fields noted 
that it is of significance. 

Up to this point we have considered the distortion of the image on 
the film itself. In certain cases this image will be projected either 
onto another film or onto a screen. When this is done, distortion of 
an entirely different nature may be encountered. It has been shown 
in a previous paper 2 that the heat of the projector lamp frequently 
causes the film to assume a curved shape during the instant the film is 
in the aperture of the projector. This curvature of the film produces 
distortion in the projected image. In the case of theater projection, 
these distortions are seldom noticeable. In special cases, where little 
or no distortion of the projected image is permissible, it will be neces- 
sary to reduce the heat intensity at the film surface with heat-absorb- 
ing media to the point at which these image distortions disappear. 

Film Defects Associated with Insufficient Drying. Although a 
great many studies have been made of most of the steps in film proc- 
essing, very little has been published on the subject of the drying 
of film. There are great differences of opinion in regard to the best 
conditions for drying. Most of the steps in the laboratory handling 
of film from printing to screening are controlled by a practical evalua- 


tion of the results obtained. For instance, if too much light is used 
in printing, the resulting print appears too dark on the screen and the 
necessary correction is made in the case of succeeding prints. If the 
processing solutions are not correct, a print of too high or too low a 
degree of contrast may result. An adjustment is, therefore, made in 
the time of processing or in the solutions. However, in the case of 



5096 R.H. 559bR.H. 

6596 RH. ( 55%R.H 





5096R.H. 8096R.H. 

90%R.H. > ) 8096R.H. 

A B 

FIG. 4. Effect of drying time on moisture 
distribution in motion picture positive film : A 
Relative humidity with which each layer is in 
equilibrium at end of drying period; B Rela- 
tive humidity with which each layer is in equilib- 
rium after standing in roll form. 

the drying of film, a step which affects the appearance and operation 
of the print for the remainder of its useful life, there is no simple guide 
for the processing superintendent. The only guide he has, the ap- 
pearance of the film, may be deceiving. Admittedly, the operator 
can see when the film is "sensibly dry," but this observation applies 
only to the emulsion layer. It tells him nothing concerning the dry- 
ness of the base. It is possible, when drying fine-grain emulsion 
films at a low dry-bulb temperature and low relative humidity, to reel 

216 R. H. TALBOT Vol 45, No. 3 

the film with the emulsion layer practically in equilibrium with the air 
of the drying cabinet, but with the base only in partial equilibrium 
with the air of the drying cabinet. This results from the fact, as 
Calhoun has pointed out, that the emulsion layer reaches equi- 
librium with air at a much more rapid rate than does the base. 
When films dried in this manner are reeled, moisture is transferred 
from the base to the emulsion until the 2 layers are in a state of 
moisture equilibrium. This effect is illustrated in Fig. 4. In these 
examples, samples of film were suspended in air at 50 per cent RH 
(relative humidity) for 2 different periods; the first was allowed to 
stay for a considerable time, and the second was removed as soon as it 
appeared "sensibly dry." 

The first sample in Fig. 4 contained 2.25 per cent moisture by 
weight, and if allowed to equilibrate without loss or gain would have 
been in equilibrium with air at 55 per cent RH. Because the emul- 
sion dries so rapidly, however, at the time the sample was removed 
for analysis the emulsion had gone substantially all the way toward 
equilibrium with the air at 50 per cent RH. Yet the base with its 
much slower drying rate had gone only 70 per cent of the way, and 
still contained enough water to be in equilibrium with air at 65 per 
cent RH. It is only after several hours' equilibration (wound in a 
roll, for example) that both emulsion and base would attain the over- 
all equilibrium of 55 per cent RH. 

The second sample, which was removed as soon as it showed the 
change in the direction of curl associated with "sensible dryness," 
contained 3.5 per cent moisture by weight, sufficient so that if allowed 
to equilibrate without further loss or gain, it would have been in equi- 
librium with air at 80 per cent RH. Nevertheless, the emulsion, at 
the time the sample was taken, had, as before, dried almost com- 
pletely to equilibrium with air at 50 per cent RH. Only in the base 
was the short drying time apparent, for here the base had had the 
opportunity to go only 20 per cent of the way toward equilibrium 
with air at 50 per cent RH. It was as if a strip of emulsion at 50 per 
cent had been cemented to a strip of base at 90 per cent; the resulting 
film on equilibration of base with emulsion eventually reaches equi- 
librium with air at 80 per cent RH. 

Even though, at the time the samples were taken, these 2 films 
looked and felt equally satisfactory, it is now apparent that only the 
first strip was sufficiently dried. The second strip with its high 
moisture content is subject to many defects directly caused by the 


moisture still present at the time of reeling. Defects often found on 
films dried in this manner are : 

(a) Tackiness of the emulsion, causing sticking in the projector gates, 

(b) High positive curl. This is especially true if the relative humidity of the 
drying cabinet air is very low and the time of drying very short. Film with high 
curl will, when wound into rolls, frequently show "spokiness" or, in the case of 
safety film, bad twist, 

(c) Buckle, caused by the edges of the film drying more rapidly than the center 
and thereby becoming shorter. This distortion will frequently cause the picture 
to appear out of focus momentarily, 

(d) Thermal "in-and-out" of focus. This is a phenomenon associated generally 
with the use of high-intensity lamps. It is aggravated by a high moisture content 
in the film. 

These film defects have been described previously. 3 
Acknowledgment. The writer wishes to express his sincere 
appreciation to Dr. E. K. Carver for his many helpful suggestions and 
for continued guidance in the preparation of the paper, and to the 
various members of the film testing departments of the Eastman 
Kodak Company for their contributions. 


1 CALHOUN, J. M.: "The Physical Properties and Dimensional Behavior of 
Motion Picture Film," /. Soc. Mot. Pict. Eng., 43, 4 (Oct., 1944), p. 227. 

2 CARVER, E. K., TALBOT, R. H., AND LOOMIS, H. A.: "Effect of High-Inten- 
sity Arcs upon 35-Mm Film Projection," /. Soc. Mot. Pict. Eng., 41, 1 (July, 1943), 
p. 69. 

3 CARVER, E. K., TALBOT, R. H., AND LOOMIS, H. A.: "Film Distortions and 
Their Effect upon Projection Quality," /. Soc. Mot. Pict. Eng., 41, 1 (July, 1943), 
p. 88. 



Summary. Some of the more important problems of theater television projec- 
tion technique and equipment are discussed. The importance of the two principles 
of optical storage and light modulation for overcoming certain limitations is stressed, 
and some Scophony developments are described in which these principles are utilized. 

It is indeed a great honor for me to discuss here before the Society 
of Motion Picture Engineers some technical aspects of television 
which are likely to become of great importance to the motion picture 

I was somewhat hesitant to give this talk because for various and 
obvious reasons I cannot enter into many technical details and cur- 
rent developments, but I realized that there were many problems of 
large screen television projection of a general nature. I intend to dis- 
cuss some of the important problems and their solutions with particu- 
lar reference to methods developed by Scophony, mainly before the 
war and perfected since. 

Let us therefore survey some of the basic requirements for theater 
television projection, and then discuss some principal limitations and 
technical approaches to overcome those limitations. 

Let us assume quite generally some sort of projection equipment for 
television, or also for motion pictures or lantern slides, consisting sub- 
stantially of an image screen surface and an optical projection system 
which forms a magnified image of this surface on the theater screen. 
The observers of this theater screen are greatly interested in the fol- 
lowing factors : 

(a) Picture or screen size, 

(&) Picture brightness, 

(c) Definition or resolution of details. 

* Presented Mar. 21, 1945, at a meeting of the Atlantic Coast Section of t he 

** Director of Research and Development, Scophony Corporation of America, 
New York. 


Further factors of interest include contrast, the ability of the system 
to reproduce fast movements, freedom from flicker, color hue of the 
picture, etc. 

We may best realize the difficulties which have to be overcome in 
television projection by comparing it with motion picture projection 
technique. This is the more important since the public will approve 
only of large screen television if the pictures supplied thereby will 
stand comparison with the motion picture projections with which 
they are familiar. 

In motion picture projection, similar to lantern slide 'projection, a 
picture of varying transparency or opacity on a transparent carrier is 
illuminated by means of a condenser system with the light of a stand- 
ard light source, which may be an incandescent or an arc lamp, and 
then projected by an optical imaging system onto the theater screen. 
Actually, at each element of the theater screen the light originating in 
the standard light source is modulated in accordance with the local 
elemental opacity values of the film or slide. The projection occurs 
simultaneously for alt elements of the picture, and in the case of mo- 
tion pictures for a substantial part of the duration of one frame. 

We shall see that the just-mentioned two facts, namely, the modu- 
lation of the light of a standard light source by the elements of the 
film picture and the simultaneous projection of all these elements to 
the screen, which are so naturally accomplished in motion picture 
projection, are not at all a priori evident in television projection tech- 
nique. However, any television system which intends to satisfy the 
public by presenting pictures comparable to motion pictures must 
necessarily aim at these two features of light modulation and of 
simultaneous projection of as many elements as possible. 

From the method of building up a television picture, it is evident 
that one cannot offhand expect a simultaneous action of all or even of 
many picture elements. A television picture is built up by the scan- 
ning process, i. e., both at the transmitter and the receiver elemental 
portions of the picture area are successively active starting, for in- 
stance, from the upper left corner, elemental portions are swept in a 
horizontal line to the upper right corner. Then jumping back to the 
left border of the picture one horizontal line lower is scanned, and so 
on until the elemental scanning spot arrives at the lower right corner, 
which completes one frame, and after which it jumps back to the 
upper left corner. With a picture of the present standard of 525 hor- 
izontal lines, repeated 30 times per sec, the time available for one 



Vol 45, No. 3 

element to be scanned is less than one five-millionth of a second. It 
was soon realized at the transmitter end that it would be necessary to 
make each picture element active for a considerably longer time 
period. This was accomplished by introduction of electrical picture 
storage, as developed by Dr. Zworykin in the Iconoscope, in which 
the light impressions and the photoelectric elemental charges are 
stored for a great part of the frame period. 

As long as one is satisfied at the receiver end with relatively small 
pictures, measuring in inches rather than in feet, one can get around 
these storage 'considerations, and build up pictures in which at any 
time substantially only one element is active. As soon, however, as 
the requirements of size and brightness become less modest, it is neces- 
sary to consider means of having as many elements as possible simul- 
taneously illuminated at the picture screen. This can be achieved by 

FIG. l. 

making use of optical storage at the receiver end and deriving there- 
from benefits as important as those derived from the electrical picture 
storage at the transmitter end. 

The term "optical storage" denotes the fact that though the optical 
changes at each picture element are created by the scanning process 
during the above-mentioned short time, each picture element retains 
its optical qualities for a time far in excess of this short elemental scan- 
ning period, in other words, "stores" the optical impressions. Thus, 
if, for instance, a given element retains its light value for a duration 
corresponding to the time in which 200 elements are successively 
scanned, this means that 200 elements are active simultaneously at 
any time. If any element retains its impressions for the whole frame 
perio'd, it means that all elements of the picture are active simul- 
taneous!^ ; this corresponds to the ideal case of motion picture pro- 

Fig. 1 illustrates these conditions showing a projection system 


throwing a picture on a picture screen P. In a slide or film projector 
all elements of the screen, or its whole surface would be simultaneously 
illuminated. In a television system without storage only one picture 
element of an elemental area representing about one two-hundred 
thousandths of the total screen area would be active at any time. In 
a television system employing a certain amount of optical storage, a 
certain part /of the screen area would be illuminated simultaneously, 
the remainder being dark. During the scanning of the picture this 
active area / will be at different parts of the screen, but the system 
may be characterized by the ratio of the active area / to the whole 
screen area F, called the storage ratio r of the system. This thus de- 
fined storage ratio would be practically zero for a system without 
storage, in which only an elemental area is active, and it would be 
practically one for a system in which all picture elements are active 

Returning now to the two most important of the above-mentioned 
factors of interest, namely, picture size and brightness, we obtain the 
following expressions : 

E = cr-I 

E = c'n^-I 

'- L r 

in which E is the screen illumination, / is the brilliance of the light 
source, a is the projection distance, d is the effective diameter of the 
projecting exit pupil, n is the number of simultaneously active picture 
elements, and c and c f are constants which include losses by absorp- 
tion, screen scattering factor, shutter action, "fly-back" time in the 
case of television, etc. 

From these simple expressions one can see that the screen bright- 
ness is proportional to the number n of simultaneously active picture 
elements, or, in other words, to the storage ratio r of the television 

Some idea of the actual values obtainable is given in Table 1. 
These values result from the above formulas by inserting therein an 
average screen brightness of 10 ft-L, using for projection an arc lamp 
of a brilliance of, say, 150,000 candles per sq in., a projection system 

222 A. H. ROSENTHAL Vol 45, No. 3 

with an effective exit pupil d of 2 in., and assuming a total loss factor 
c of 50 per cent. 




T a/d Projection 

Simultaneous Storage Projection Distance, 

Elements Ratio Ratio f or d 2 in. 

1 1/200,000 4 8 in. 

300 1/700 70 12ft 

100,000 1/2 1300 200 ft 

200,000 1 1800 300ft 

The first column shows the number of simultaneously active ele- 
ments (with a 525-line standard), the second column the correspond- 
ing storage ratio, the third column the projection ratio, that is, pro- 
jection distance a to projection opening d, and the fourth column 
shows the maximum projection distances a max allowable in order to 
obtain the above screen illumination of 10 ft-L. This table, the values 
of which are based on average motion picture theater conditions, 
shows that for a television system in which only one element is op- 
tically active at a time, that is, a system with no storage, a very short 
projection distance of less than one foot results. It is obvious that 
even with a projection system of extreme wide-angle type only a pic- 
ture size of the order of the projection distance could be obtained, that 
is, of less than one-foot width. 

Thus the values show the impossibility of obtaining even with a 
high-intensity arc lamp a picture of reasonable size and brightness 
with a television projection system with no storage. 

The second row, which represents a television system in which 300 
elements are simultaneously active, gives a maximum projection dis- 
tance of 12 ft, and thus could provide a picture of average theater 
screen brightness and of a linear size of the order of 12 ft. Naturally, 
the assumption of an effective opening of 2 in. was made with a view 
to motion picture projection systems, and the conditions for a tele- 
vision system of limited storage will look more favorable if a projec- 
tion system with a larger aperture is used. 

But it is obvious that one soon arrives at a practical limit, even 
with special projection systems. For instance, in the first case of a 
television system without storage, in order to obtain a theater picture 
of about 20-ft width one would require a projection system of an 
effective aperture of at least 60 in. diameter, which seems to be some- 
what on the impractical side. 


The next 2 rows show corresponding values for storage ratios of 
one-half and of one. Thus with a television system in which half of 
the picture elements are simultaneously active, one would obtain a 
maximum projection distance of about 200 ft, which of course would 
permit a picture of very large size, sufficient even for the largest 
theater screens such as, for instance, in drive-in theaters. With a 
projection distance of 100 ft either the effective diameter of the pro- 
jection system could be reduced to about one inch, or with a 2-in. 
system a very large screen brightness of approximately 40 ft-L could 
be obtained. 

These figures are only intended to illustrate the great importance of 
optical storage in television projection systems, that is, of designing 
the system in such a way that as many as possible of the picture ele- 
ments are simultaneously illuminated on the screen. Only by a high 
optical storage can a television picture of satisfactory size and bright- 
ness be produced with economical optical means. 

I have just tried to illuminate the great advantages of a television 
projection system with a high optical storage ratio by comparing it 
with a system with no storage. Similarly, we can appreciate the ad- 
vantages of the principle of light modulation by discussing the limita- 
tions of television projection systems in which no modulation of the 
light of a standard light source is used. 

The most prominent, and probably the only practical example of 
such a television system using no standard light source and modulat- 
ing the light thereof, is based on the fluorescent screen cathode-ray 
tube. Here the picture is produced on a fluorescent screen by scan- 
ning the screen with an intensity modulated cathode-ray beam. The 
electric energy of this modulated cathode-ray beam when impinging 
the fluorescent material is converted, at least in part, into light energy, 
and a self -luminescent picture thus created on the fluorescent screen. 
The image can be observed directly if one is satisfied with picture 
sizes that can be accommodated on the end face of a cathode-ray 
tube, and which for practical reasons are limited to the order of 10 in. 
in width. 

When larger pictures are desired, it is necessary to project this self- 
luminescent picture by an optical system onto the large viewing 
screen. The principal disadvantages of this method become apparent 
if one aims at pictures of theater size and theater brightness. Size 
and brightness of the picture determine the total light energy which 
must issue from the exit pupil of the projection system. Whereas in 

224 A. H. ROSENTHAL Vol 45, No. 3 

light modulating projection systems, whether for television or for film 
projection, this light energy is derived from a strong standard light 
source and is quite independent of the modulation and the television 
signals themselves; in the case of a self -luminescent picture, the light 
energy at any moment is determined by the modulated electric energy 
of the cathode-ray beam which is converted into light energy at the 
fluorescent screen. Thus in a fluorescent cathode-ray tube the total 
light output represents the converted modulated power of the cath- 
ode-ray beam whereby the beam modulation is produced by the am- 
plified television signals. Further, this light energy is being created 
at only one picture element at a time. 

With a 525-line picture, the cathode-ray beam energy which is far 
in excess of the total light energy obtainable must be concentrated in 
a spot of 1 /2oo,ooo of the area of the fluorescent screen which, in the case 
of a 10-in. screen, amounts to about one-tenth of a square millimeter. 

Since the light output of a fluorescent screen is only up to a certain 
beam current density proportional to the beam current, and satura- 
tion occurs already at a current density considerably below that re- 
quired for satisfactory brightness and definition, the total light output 
can be practically increased only by increasing the spot size or, in 
other words, by sacrificing picture definition. 

Further, a fluorescent screen represents a rather perfectly diffusing 
light source. An optical system with an aperture f/2 would gather 
only about 6 per cent of the light flux emitted in a forward direction by 
an axial element of the fluorescent screen. In order to gather an 
appreciable amount of the limited light flux of the self -luminescent 
picture, special optical systems of very high aperture, such as the 
Schmidt system, have to be used. But even then, with the extremely 
high voltages of about 70,000 v necessary with these projection 
cathode-ray tubes, it appears doubtful whether a practical theater 
television projection system giving pictures of satisfactory size, bright- 
ness, and definition will ultimately be based on the self -luminescent 
fluorescent cathode-ray tube method. 

A projection method in which the light energy is not derived from 
converted signal modulated power, but from a standard light source, 
and in which the signals serve only to modulate in the way of a relay 
or valve the intensity of this standard light source, is inherently free 
from these limitations. Conventional optical systems can be used to 
concentrate the light from the standard light source upon the light 
modulating device, and to project therefrom a picture on the screen 


by means of directed light similar to film projection without a great 
deal of loss, as is the case of the perfectly diffusing self -luminescent 
fluorescent picture. 

Further, provided a sufficient amount of optical storage and light 
flux directed to the viewing screen is available in the system, a prac- 
tically complete independence of picture brightness and definition 
results, the brightness being determined by the light source and opti- 
cal storage, and the definition being determined independently by 
characteristics of the light valve. 

1 shall discuss the light valve 
later in connection with typical 
examples for light modulating 

Summarizing, we have seen 
the increasing importance of the 
two principles of optical storage 
and of light modulation as we 
approach conditions of theater 
television projection. 

We have been interested in 
theater television because of ^ur 
close commercial connections 
with motion picture theater 

Rather than forcing improve- 
ments on existing systems which 
inherently appear to be not so 
well adapted to the large screen 
problems, and overstraining their 
possibilities, we aimed at developing systems which would be free 
from such limitations and would therefore offer greater scope for 
any possible increased future requirements as to picture standards. 
Therefore the efforts of the research and development work were 
directed toward systems employing the principles of storage and light 

By way of example, I should like to describe briefly the following 

2 systems which, though based on different physical phenomena, both 
employ these two principles. 

The first is the Supersonic system. It is based on the diffraction 
of light by supersonic waves in a liquid. The liquid is contained in 

.Fie. 2. 

Supersonic light modulator 

226 A. H. ROSENTHAL Vol 45, No. 3 

the supersonic light modulator cell (Fig. 2) which forms an essential 
part of this system. Supersonic traveling waves are created in the 
liquid by a piezoelectric quartz plate in contact therewith and excited 
by an electric oscillator to supersonic resonance vibrations of a fre- 
quency between 10 and 20 megacycles. The electric oscillations are 
amplitude-modulated by the television signals. Thereby supersonic 
waves of varying intensity are propagated from the crystal to travel 
in the liquid column. Light beams from a standard light source are 
sent through this liquid column. Depending upon the local ampli- 
tude of these waves, more or less light is diffracted from its path be- 


FIG. 3. 

cause supersonic waves represent periodic compressions and rarefac- 
tions of the liquid, resulting in periodic variations of its refractive in- 
dex. These optical variations, which are arranged along the liquid 
column, act as a variable diffraction grating upon the light beams. 
This can be seen from Fig. 3. Light is concentrated by condenser 
lenses upon a slit and is made parallel by a further lens to traverse the 
supersonic cell. A diffraction spectrum results in an image plane if 
supersonic waves are excited in the cell. 

Fig. 4 shows that the diffraction spectra are produced in an image 
plane of the illuminating slit. A bar or stop is arranged in this plane 
so that it will just obscure this image if no supersonic waves are pres- 
ent. Thus no light will pass this stop. If, however, supersonic waves 
are excited in the cell, light is diffracted from its normal path and will 


therefore pass the stop in proportion to the intensity of the supersonic 
waves, or of the television signals. Thus the device acts as a light 

It is very interesting and useful that the supersonic waves excited 
by the signal modulation carry this modulation along in their propa- 
gation through the liquid. It is as if a blank film would pass with a 
certain velocity along the crystal at which point modulations are im- 
pressed on the film which are carried along by it. Thus the various 
successive parts of the film, or of the liquid column, will carry modu- 
lations corresponding to successive picture elements of the television 
picture. Depending upon the length of the liquid column, a consider- 




FIG. 4. 

able number of successive picture elements are thus impressed upon 
this column at any moment and can be utilized. The supersonic 
waves travel with the sound velocity of the liquid, which is of the 
order of three-quarters of a mile per second. Since a picture element 
of a 525-line picture corresponds to about one five-millionths of a sec- 
ond, the length of the liquid column carrying the modulation of one 
element corresponds to about one-fifth of a millimeter in the cell. 
Thus an active liquid column of 2 in. would accommodate about 250 
picture elements. Since a column of this or even a greater length is 
utilized, it can be seen that the supersonic light modulator permits a 
storage or simultaneous representation of several hundred picture 
The liquid column, itself, with its impressed picture modulations is 



Vol 45, No. 3 

imaged upon the viewing screen, and represents thereon a considerable 
part of a picture line. Since these modulations, however, move in that 
column with sound velocity, the picture elements also move in the line 
image represented on the screen. It is necessary to stop this move- 
ment by introducing an opposite compensating movement. That is 
accomplished by a rotating mirror polygon called the high-speed 
scanner, which immobilizes the elements on the screen. Fig. 5 shows 
this scanner introduced behind the bar. 

FIG. 5. 

It is still necessary to introduce a deflection perpendicular to the 
lines in order to arrange the consecutive picture lines spaced properly 
in a 2-dimensional picture. This is accomplished by a second rotating 
mirror polygon called the low-speed scanner. Fig. 6 shows the com- 
plete principal arrangement. This figure also shows several cylindri- 
cal lenses which are quite important to realize in a practical and 
economical way the advantages of the system, and which make pos- 
sible a great reduction in the size of the high-speed scanner which has 
to rotate with a very high speed. The rotation of the scanners is con- 
trolled by the synchronizing signals of the television system, and very 



effective motors have been developed for this purpose. The system 
thus employs the essential principles of light modulation and optical 
storage. For theater purposes, a standard motion picture arc lamp 
is used as light source, and Fig. 7 shows a complete projector for 
theater use, in which the arc lamp can be recognized. 

Television pictures of theater size have been shown regularly and 
commercially in London theaters and also experimentally in a New 
York theater before the war. To give an impression of the great pub- 
lic interest shown, and to be expected for theater television, Fig. 8 

shows a crowd of people in front of one of the London theaters in 
which this equipment was used. 

The next example concerns another development based on quite 
different physical principles. 

I had mentioned in the beginning that in terms of television tech- 
nique, motion picture or slide projection can be regarded as a light 
modulation process with all elemental portions simultaneously active, 
thus a light modulation system with full storage. 

A television system which would most closely resemble motion pic- 
ture or slide projection technique would contain a small screen of the 
size of a lantern slide, for instance, the elemental transparencies of 



Vol 45, No. 3 

which could be controlled in a simple way by the television signals. 
Retaining the local elemental transparencies for a considerable part 
of the frame period would result in a high storage ratio. Such a screen 
of variable transparency could then be projected similar to a lantern 
slide or film, using a standard light source. 

A method using this principle is the Skiatron. It consists substan- 
tially of a special type cathode-ray tube in which the self -luminescent 
fluorescent screen is replaced by a screen of a material exhibiting the 

property of "electron opacity," 
that is, the normally transpar- 

JB| en "t material can be rendered 

^1 & more or less opaque by the 

i^ electrons of an impinging cath- 

ode-ray beam. This inter- 
esting electron opacity effect 
which occurs in various ionic 
crystal materials, for instance 
of the alkali-halide class, has 
been for many years a scien- 
tific curiosity, until the writer 
realized that it happens to 
have very desirable qualities 
for television and other ap- 
plications. This was con- 
firmed by our experimental 
and theoretical investigations. 
The physical nature of the 
electron opacity is quite com- 
plicated and by no means 
completely explored as yet, 
similar to the nature of fluo- 

FIG. 7. 

Scophony theater television pro- 

rescence to which it is related. But we can get some idea of what 
happens with a simplified model as shown in Fig. 9. 

Such an ionic crystal, for instance, potassium chloride, consists of a 
lattice of alternately arranged positive potassium ions and negative 
chlorine ions. Suppose that a layer of such a crystal is subjected to 
an electric field as indicated. The crystal normally is transparent to 
visible light. If, now, an electron from a cathode-ray beam enters the 
crystal lattice, it will be attracted by the nearest positive potassium 
ion, and will form with it some sort of loosely bound potassium atom 


which now shows an absorption in the visible spectral range, and is 
therefore called a color center. 

The binding forces between the electron and the ion are rather 
weak, and after a short time the electron is again split from its ion by 
the thermal vibration forces of the crystal lattice. The electron 
moves on in the electric field until it is again captured by another 
positive potassium ion, forming again a color center a little nearer to 
the anode of the electric field. This process is repeated many times, 
the electron being, so to say, made visible temporarily when it is 

FIG. 8. 

bound as a color center to lattice ions. Exchanging from ion to ion 
it finally approaches the anode side of the electric field and leaves the 

Figs. 10 to 15 show these opacities in an ionic crystal. A small 
potassium bromide crystal in this experiment is held in a frame and 
at a suitable temperature, and a platinum point inserted at one side, 
while the opposite side has a flat metal electrode. When the plati- 
num point is given a sufficiently negative potential with respect to 
the other electrode, electrons enter from the point into the crystal 
and are made visible in the manner described as an opaque cloud. 
The various photographs of this crystal were made at different times 
(Fig. 10 before applying the potential) and show how the opacity 



Vol 45, No. 3 

develops in the crystal. When the field is reversed (Figs. 14 and 15), 
the opacity cloud returns again toward its source and leaves the 
crystal clear. This experiment is somewhat different from the tele- 
vision application, but illustrates the nature of the electron opacity. 
Such experiments are useful to measure various factors of interest 
for the applications. 

If a cathode-ray beam containing many electrons impinges for a 
certain time on a given area of such a crystal layer, a local opacity is 
created at that area, and this opacity, representing electrons, moves 
across the crystal layer and disappears after a given time leaving the 

e $ e 
e e 
e e 
e e 


e e e 
e e 
a e e 
e e 
e e 
e e 
e e 

' el. f, 

FIG. 9. 

crystal layer clear. A closer investigation of this electron opacity 
phenomenon shows that the speed of movement and disappearance 
of the opacity depend upon various factors, such as the electric field, 
temperature, nature and thickness of the crystal material, and that 
it can be influenced within wide limits by various factors. Of greatest 
interest to our problem is the fact that these magnitudes are within 
ranges suitable for television projection. 

Fig. 16 shows a schematic diagram of such a Skiatron projection 
arrangement in which you see the tube with its crystal screen in which 
the fugitive opacity image is created by scanning this screen with the 
signal modulated cathode-ray beam. This fugitive image consists of 
areas of variable opacity or transparency, quite similar to a film or 
lantern slide picture, and it can be projected in a similar way by a 


standard light source, condenser, and projection lens onto a projec- 
tion screen through suitable windows in the tube wall. We have thus 
a simple shadow picture projection, which is the basis for the name 
Skiatron from the Greek word skia meaning shadow. 

Generally speaking, any intelligence traced upon the electron 
opacity screen by a cathode-ray beam is represented thereon by tem- 
porary local changes in its optical properties, such as its transparency, 
its reflective power, its refractive index, etc. Therefore, these traces 
can be made visible in various ways by illuminating the screen (with 
various light sources, also daylight), apart from the just-described 
projection method, which is only one of the various applications em- 
bodied in the Skiatron patent group. 

A high storage effect is inherent in this method by substantially 
retaining the opacity values unchanged over the frame period. A 
previously created opacity in a given area can be made to move 
across the crystal within the frame period, thus remaining substan- 
tially constant, and to disappear at the end of this period, when the 
scanning beam returns to the same area, inserting therein the electrons 
and creating the opacity for the next frame. 

Fig. 17 shows a laboratory setup of the system in which can be 
recognized the. tube, the condenser, and the projecting lens. The 
electron opacity screen is of microcrystalline structure. 

Experiments in the course of developing this system, including tele- 
vision picture reception, have shown great promise that the above- 
mentioned ideal performance, which can be anticipated from the 
physical basis of these effects, should be fully realized by continued 
research dealing mainly with the electron opacity effects. It has been 
mentioned that these effects are physically related to fluorescence and 
phosphorescence, and contemplating the great improvements of fluo- 
rescent materials from a few years ago when they were first applied in 
television tubes, one can anticipate similar important developments 
of this scientifically related art which appears to be far more adapted 
to the problems of television projection. 

Since the light output in the projected picture is derived from a 
standard light source and does not represent converted modulated 
power, and thus the amplified television signals are used only to 
modulate and not to generate the local light intensities, it is not 
necessary to sacrifice picture definition for picture brightness here. 
The definition is determined only by electron optical questions which 
are well solved in modern cathode-ray tube technique. Experiments 


A. H. ROSENTHAL Vol 45, No. 3 

FIG. 10. 

FIG. 11. 

FIG. 12. 


FIG. 13. 

FIG. 14. 

FIG. 15. 



Vol 45, No. 3 

indicate that even a far greater number of lines than the present 525 
can be used. Also the Supersonic system, as will be clear from pre- 
vious description, is not limited to present definition standards. 

The Skiatron represents a cathode-ray controlled light valve and 
thus a purely electronic system. It employs the principles of light 
modulation and high optical storage, and there is much reason to 
expect that this system may ultimately acquire great importance 
both for theater and home television. 

The examples described above of theater television methods are in 
many respects similar to motion picture technique. There comes to 

Projection screen Skiatron tube 
Projection lens 


^ Light source 

Cathode my 


mind another theater television projection method, namely, the inter- 
mediate film process, in which television pictures produced on a 
cathode-ray tube screen are photographed on a motion picture film, 
and, after speedy processing of this film, projected through a motion 
picture projector. It is interesting to note that the Skiatron method, 
producing pictures of varying transparency, resembles in this respect 
the intermediate film process without, however, consuming film. 
There is even a deeper physical analogy between the two methods be- 
cause the electron opacity effects are basically related to the photo- 
graphic process. 

Among the various embodiments of the Skiatron principle there is 
one in which the electron opacity pictures are created by the scanning 


and modulated cathode-ray beam on a movable, endless carrier, such 
as a film loop or disk carrying the ionic crystal material. At one posi- 
tion of this carrier the electron opacity image is impressed thereon, 
then the carrier traverses a projector in which the image is projected 
to the viewing screen, at a later position of the carrier the picture is 
extinguished, and then returning to the first position the carrier re- 
ceives the opacity picture of the next frame. 

May I briefly say a few words about color television projection? 
Small color television pictures were demonstrated both in this country 

I v / 

V X ^a. 

FIG. 17. 

and in England before the war, by making use of a color filter disk ro- 
tating in front of the picture on a cathode-ray tube, and thus succes- 
sively presenting to the eye of the observer partial color pictures. 
This disk rotating in front of the picture has to be quite big. 

Such an additive successive color method can be easily applied to 
the Skiatron and Supersonic methods, and in this case the filter disk 
can be made quite small by inserting it near the optical imaging sys- 
tem where the whole light energy is restricted to a small cross section. 

The Skiatron method can also provide a sub tractive color tele- 
vision system. This is based on the fact that the color centers or 
opacities have different colors for different crystal materials. Project- 
ing the light through 3 electron opacity screens Si, S 2 , S 3 of suitable 



Vol 45, No. 3 

materials, and creating on each of these screens an opacity picture 
corresponding to one partial picture of a 3 -color system (thus of 

FIG. 18. 

minus red, minus green, and minus blue colors), leads to a subtractive 
color television system. The principle of this is shown in Fig. 18, in 
which you see 3 Skiatron tubes 7\, T 2 , T 3 through which light is pro- 

FIG. 19. 

jected from a standard light source L to a viewing screen P by lenses 
0i, 02, 03, taking care that the partial pictures are projected in regis- 
ter. Two of the screens may also be combined in one tube. 



From the experiences of color photography and color motion 
pictures, the far greater optical efficiency of a subtractive color pro- 
jection process compared with an additive one is well known. Pres- 
ent color motion picture and photography processes are therefore 
based on subtractive methods, like Technicolor and Kodachrome. 
The same considerations apply to projection of color television, and it 
may be anticipated that here also a subtractive process of the kind de- 
scribed will finally be used. 

FIG. 20. 

A further aspect of a full storage system, that is, of a continuous 
melting of successive picture frames, to which the Skiatron method 
lends itself, is the possibility of reducing the frame repetition frequency 
without causing undue flicker and impairing motions. With full 
storage this frequency could be reduced by about 50 per cent, there- 
with also similarly reducing the total frequency band width of the 
television signal spectrum. This should be very important in view 
of the discussions aiming at increasing the number of lines, particu- 
larly for theater television, ultimately to perhaps 1000, which 
would greatly increase the necessary frequency band. 

I have tried to sketch some of the more important problems of 
theater television equipment, and to illustrate these problems, by 
way of example, by briefly describing the principles of some Scophony 


developments, rather than going into technical details. I may just 
mention that many far-reaching improvements have been and are 
being made on these methods beyond what has already been pub- 
lished previously. 

Finally, though it is not within the scope of this talk which deals 
with theater television, it may be mentioned that the previously de- 
scribed methods based on light modulation and optical storage can 
be very successfully applied to smaller pictures, in which case the 
carbon arc is replaced by smaller light sources. To illustrate this, 
Fig. 19 shows a Scophony projector giving a picture of 4-ft width 
which has been designed for schools and lecture halls. Fig. 20 shows 
2 models of Scophony home projection receivers giving pictures of 18 
and 24 in., respectively. 

When television is generally available to the public, public reaction 
will provide, as it has in radio, many new problems for the research 
laboratories. But the public should from^the start be far better off 
than the students of the piano teacher who put in his window a sign 
reading: "Piano Lessons Special Pains Given to Beginners." 


(1) ROSENTHAL, A. H. : Proc. I. R. E., 28 (1940), p. 203. 

(2) ROBINSON, D. M.: Proc. 1. R. E. t 27 (1939), p. 483, and following articles 
on Scophony developments. 

(3) ROSENTHAL, A. H.: Electronics & Television, 13 (Feb.-Mar., 1940). 

(4) ROSENTHAL, A. H. : Electronic Eng. t 14 (1942), p. 578. 



The editors present for convenient reference a list of articles dealing with subjects 
cognate to motion picture engineering published in a number of selected journals. 
Photostatic or microfilm copies of articles in magazines that are available may be 
obtained from The Library of Congress, Washington, D. C., or from the New York 
Public Library, New York, N. Y., at prevailing rates. 

American Cinematographer 

26 (July, 1945), No. 7 

The Museum of Modern Art Film Library (p. 226) 
Rerecording 35-Mm Entertainment Films for 16-Mm 
Armed Forces Release (p. 232) 

Electronic Industries 

4 (Aug., 1945), No. 8 
Television Optics (p. 80) 


18 (July, 1945), No. 7 
Supersonic Bias for Magnetic Recording (p. 126) 

International Photographer 

17 (June, 1945), No. 5 
The Adel Color Camera (p. 22) 

International Projectionist 

20 (June, 1945), No. 6 

Operation and Maintenance of the Filmosound V 16-Mm 
Projector (p. 7) 

Sound Reduction Amplifiers (p. 10) 

Motion Picture Projection in the Soviet Union (p. 13) 

Swiss Television System Combines Electron Beam and 
Arc Lamp (p. 14) 

Projectionists' Course on Basic Radio and Television 
Pt. 12 (p. 18) 

New Continuous Projection System Invented by Cana- 
dian (p. 24) 

20 (July, 1945), No. 7 

A Step-by-Step Analysis of the Filmosound V 16-Mm 
Amplifier (p. 7) 

The Projection of Thomascolor Motion Pictures (p. 12) 

Sound Failure at Fine- Wire Contacts (p. 14) 











Projectionists' Course on Basic Radio and Television 

Pt. 13 (p. 18) M. BERINSKY 

DuMont's Projection Tele (p. 22) 

Radio News 

34 (July, 1945), No. 1 
Color Television (p. 32) R. W. EHRLICH 


2 (June, 1945), No. 5 
Post-War DuMont Cameras (p. 11) H. T. TAYLOR, JR. 


Hotel Pennsylvania, New York 
October 15-17, 1945 

Directory of Committee Chairmen 

Atlantic Coast Section and Local Ar- 
rangements C. R. KEITH, Chairman 

Papers Committee BARTON KREUZER, Chairman 

C. R. DAILY, Vice-Chairman 

Publicity Committee HAROLD DESFOR, Chairman 

Registration and Information W. C. KUNZMANN 

Membership and Subscription Com- 
mittee JAMES FRANK, JR., Chairman, East Coast 

"Victory" Dinner-Dance D. E HYNDMAN 

Hotel and Transportation O. F. NEU 

Projection Programs 35-mm H. F. HEIDEGGER, Chairman, assisted by 

Members New York Projectionists 
Local 306 
16-mm M. W. PALMER 


No room reservation cards will be mailed to the membership either by the hotel 
or the Society and those who are contemplating attending this Conference should 
book their room accommodations direct with D. M. Mumford, Resident Manager, 
Hotel Pennsylvania, New York 1, New York, immediately. No rooms will be as- 
sured or available unless confirmed by the hotel management. 

The following per diem room rates, European plan, are extended to SMPE 
members and guests when booking reservations direct with the hotel: 

Room with bath, one person $3 . 85-$7 . 70 

Room with bath, two persons, double bed 5. 50- 8. 80 

Room with bath, two persons, twin beds 6. 60- 9.90 

Parlor suites living room, bedroom, and bath $10.00, $11.00, $13.00 and 





The Conference registration headquarters will be located on the 18th floor of the 
hotel adjacent to the Salle Moderne, where all technical and business sessions will 
be held during the 3-day Conference. Members and guests are expected to regis- 
ter. The fee is used to defray Conference expenses. 


Authors contemplating presentation of papers at this Technical Conference 
should submit a complete manuscript to the Papers Committee not later than 
October 1 for listing in the Final Program. 

Only through your cooperation can the Papers Committee arrange a suitable 


An informal "Victory" Dinner-Dance, Journal Award presentation, distribu- 
tion of Fellow Membership certificates to 1945 Fellows-elect, and a social get-to- 
gether will be held in the Georgian Room of hotel, on Tuesday evening, October 
16 (dress optional). 

Tickets for this function should be procured at the registration desk not later 
than noon on October 16, so that hotel accommodations may be provided accord- 
ingly. Your cooperation is solicited. 


There will be no official ladies' entertainment committee, or any prearranged 
program. However, the ladies will be welcome to attend any sessions of interest 
and the Dinner-Dance on October 16. Conference identification cards will be 
available to the ladies which will be honored at deluxe motion picture theaters in 
New York during the 3-day Conference. Application for these cards should be 
made at registration headquarters. 


Conference identification cards issued to registered members and guests will 
be honored at deluxe motion picture theaters in New York, which will be listed on 
the back of these cards. 


Monday, October 15, 1945 

Open Morning. 

10:00 a.m. Hotel, 18th Floor: Registration. Advance sale of "Victory" Dinner- 
Dance tickets. 

2:00 p.m. Salle Moderne: Opening session of Conference. 
8: 00 p.m. Salle Moderne: Evening Session. 


Tuesday, October 16, 1945 

10: 00 a.m. Hotel, 18th Floor: Registration. Advance sale of "Victory" Dinner- 
Dance tickets. Tickets must be obtained before noon to insure 

Salle Moderne: Morning Session. 
2: 00 p.m. Salle Moderne: Afternoon Session. 

8: 00 p.m. Georgian Room: "Victory" Dinner-Dance. Social get-together, 
dancing and entertainment. 

Wednesday, October 17, 1945 

Open Morning. 

2: 00 p.m. Salle Moderne: Afternoon Session. 
8: 00 p.m. Salle Moderne: Evening Session. Adjournment of the Fifty-Eighth 

Semi-Annual Technical Conference. 
All technical sessions will open with an interesting 35-mm motion picture short. 


Those desiring hotel rooms prior to and during the Conference should book 
their accommodations direct with the Hotel Pennsylvania management immedi- 
ately. No rooms will be available at this hotel unless confirmed by them. 


Convention Vice-P resident 



At a meeting of the Board of Governors held in New York on July 19, 1945, it 
was unanimously resolved to submit the following proposed amendments of the 
By-Laws to the membership of the Society for voting during the Fifty-Eighth 
Semi-Annual Technical Conference in New York, October 15-17, 1945, inclusive. 
Proposed changes are indicated in italics. 

Proposed Amendment of By-Law /, Sec. 3 (b) 

"Fellow membership may be granted upon recommendation of the Fellow 
Membership Award Committee, when confirmed by a three-fourths majority 
vote of the Board of Governors. Nominations for Fellow shall be made from the 
Active membership." 

Proposed Amendment of By-Law VII, Sec. 1 (5th paragraph) 

"The voter shall then indicate on the ballot one choice for each office, seal the 
ballot in the blank envelope, place this in the envelope addressed to the Secre- 
tary, sign his name and address on the latter, and mail it in accordance with the 
instructions printed on the ballot. No marks of any kind except those above 
prescribed shall be placed upon the ballots or envelopes. Voting shall close 
seven days before the opening session of the annual fall convention." 

Proposed Amendment of By-Law XI, Sec. 6 (1st and 5th paragraphs) 

"The officers and managers of a Section shall be Active, Fellow, or Honorary 
members of the General Society. All officers and managers shall be elected to 
their respective offices by a majority of ballots cast by the Active, Fellow, and Honor- 
ary members residing in the geographical area covered by the Section." 

"The voter shall then indicate on the ballot one choice for each office, seal 
the ballot in the blank envelope, place this in the envelope addressed to the 
Secretary-Treasurer, sign his name and address on the latter, and mail it hi ac- 
cordance with the instructions printed on the ballot. No marks of any kind 
except those above prescribed shall be placed upon the ballots or envelopes. 
Voting shall close seven days before the opening session of the annual fall conven- 





Position open for man or woman with experience in optical instrument 
design. Position also open for man or woman with experience in lens 
design or computing. Write for interview. Binswanger and Company, 
Optics Division, 645 Union Ave., Memphis, Tenn. 

Physicist with special training in optics for research on utilization of 
carbon arcs particularly in projection systems. Apply to Research Labo- 
ratory, National Carbon Co., Inc., P. O. Box 6087, Cleveland 1, Ohio. 

Designer and engineer experienced in optics, lighting, and microphotog- 
raphy, capable of designing microfilm reading equipment and products 
related to microfilm industry. Reply to Microstat Corporation, 18 
West 48th St., New York 19, N.Y. 

Design engineer, experienced in mechanics and optics of motion picture 
cameras, projectors, and film scanning. Give details. Reply to Mr. 
John H. Martin, Columbia Broadcasting System, Inc., 485 Madison 
Ave., New York 22, N.Y. 


Sound recording engineer, 16- or 3 5 -mm equipment, studio or location 
work, single or double system. Free to travel. For details write J. J. K., 
354 Ninth Ave., New York 1, N.Y. 

Honorably discharged veteran with 15 years' experience in all phases of 
motion picture production, including film editing, directing, producing. For 
details write F. A., 30-71 34th St., Long Island City 3, N.Y. Telephone 
AStoria 8-0714. 

Projectionist-newsreel editor with 15 years' experience just released 
from service. Willing to locate anywhere. Write P. O. Box 152, Hamp- 
den Station, Baltimore 11, Maryland. 

Director of visual training aids, now concluding government work, de- 
sires position with educational, industrial or commercial organization, 
supervising or assisting in the production or distribution of visual training 
or allied work. Write H. C. B., 348 Maryland Ave., Dayton 4, Ohio. 








Vol45 OCTOBER, 1945 No. 4 


Reports of SMPE Committees : 

Studio Lighting Committee 249 

Standards Committee 261 

Subcommittee on Screen Brightness 262 

Subcommittee on Theater Engineering Construction, 
and Operation 262 

Two New Eastman Fine-Grain Sound Recording Films 

Du Pont Fine-Grain Sound Films Types 232 and 236 

H. W. MOYSE 285 

The U. S. Naval Photographic Services Depot 

F. M. HEARON 294 

An Optical Cueing Device for Disk Playback 

G. C. MISENER 297 

Development of Two Automatic Follow- Focus Devices 
for Use in Cinematography 


Current Literature 310 

Copyrighted, 1945, by the Society of Motion Picture Engineers, Inc. Permission to republish 
material from the JOURNAL must be obtained in writing from the General Office of the Society. 
The Society is not responsible for statements of authors or contributors. 

Indexes to the semi-annual volumes of the JOURNAL are published in the June and December 
issues. The contents are also indexed in the Industrial Arts Index available in public libraries. 





Board of Editors 





Officers of the Society 
**President: DONALD E. HYNDMAN, 

360 Madison Ave., New York 17. 
**Past-Presidcnt: HERBERT GRIFFIN, 

133 E. Santa Anita Ave., Burbank, Calif. 
** Executive Vice-President: LORBN L. RYDER, 

6451 Marathon St., Hollywood 38. 
* Engineering Vice-President: JOHN A. MAURER, 

37-01 31st St., Long Island City 1, N. Y. 
** Editorial Vice-President: ARTHUR C. DOWNES, 

Box 6087, Cleveland 1, Ohio. 
* Financial Vice-President: ARTHUR S. DICKINSON 

28 West 44th St., New York 18. 
** Convention Vice-President: WILLIAM C. KUNZMANN, 

Box 6087, Cleveland 1, Ohio. 
* Secretary: E. ALLAN WILLIFORD, 
230 Park Ave., New York 17. 
*Treasurer: M. R, BOYER, 
350 Fifth Ave., New York 1. 


*FRANK E. CARLSON, Nela Park, Cleveland 12, Ohio. 
**JOHN I. CRABTREB, Kodak Park, Rochester 4, N. Y. 
**CHARLBS R. DAILY, 6451 Marathon St., Hollywood 38. 

*EDWARD M. HONAN, 6601 Romaine St., Hollywood 38. 
*tCLYDB R. KEITH, 233 Broadway, New York 7. 

*G. T. LORANCE, 92 Gold St., New York 7. 
**PBTER MOLE, 941 N. Sycamore Ave., Hollywood. 
*fHoLLis W. MOYSE, 6656 Santa Monica Blvd., Hollywood. 
** WILLIAM A. MUELLER, 4000 W. Olive Ave., Burbank, Calif. 

*EARL I. SPONABLB, 460 West 54th St., New York 19. 
**REEVB O. STROCK, 111 Eighth Ave., New York 11. 

*WALLACB V. WOLFE, 1016 N. Sycamore St., Hollywood. 

Term expires December 31, 1945. fChalrman, Pacific Coast Section. 
**Term expires December 31, 1946. {Chairman, Atlantic Coast Section. 

Subscription to nonmembers, J8.00 per annum; to members, $5.00 per annum, included in 
their annual membership dues; single copies, SI. 00. A discount on subscription or single copies 
of 15 per cent is allowed to accredited agencies. Order from the Society at address above. 
Published monthly at Easton, Pa., by the Society of Motion Picture Engineers, Inc. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 

General and Editorial Office, Hotel Pennsylvania, New York 1, N. Y. 

Entered as second-class matter January 15, 1930, at the Post Office at Easton, 

Pa., under the Act of March 3, 1879. 


Vol 45 OCTOBER, 1945 No. 4 


Summary. This report deals with operation and maintenance of studio lighting 
equipment. A previous paper 1 listed all of the various types of studio lighting equip- 
ment and gave data on the sizes and characteristics of carbon trims and incandescent 

Illustrations and tables show the effects of proper and improper positioning of 
carbons in high-intensity arc units. Light output and color temperatures versus 
burning hours of incandescent bulbs are illustrated. The value of the use of tungsten 
powder as a cleaning agent in motion picture studio bulbs is shown. The effects of 
varying line voltages on studio light sources are described and illustrated. 


The Rotating High-Intensity Arc. In the high-intensity arc 
(Fig. 1) the positive carbon consists of a carbon shell or tube con- 
taining a core of a mixture of carbon and certain other substances 
which, under the action of the arc, volatilize and form a ball of ex- 
tremely brilliant gases within a crater that is formed at the tip. The 
negative carbon also has a very small core which serves principally 
to keep the negative flame centered at the tip of the negative 

Feed Control. Studio-type high-intensity arc lamps are semi- 
automatic. When the lamps are in operation the feed motors are 
energized by current at the arc voltage. Therefore, if the line 
voltage drops, the current through the carbons will also drop reducing 
the carbon burning rate, and the carbons will tend to feed together. 
But as they approach each other the arc becomes shortened, the arc 
voltage drops and since the motor is energized at arc voltage, it will 
rotate slower on its reduced voltage and lower the carbon feed rate. 
In this manner a balance is automatically attained between carbon 
burning rate, arc voltage, feed motor speed, and carbon feed rate. 

* Submitted June 7, 1945. 




Vol 45, No. 4 

Once the arc control rheostat is set for proper feed rate and correct 
arc position is established very little manual adjustment is required. 

Feed Control Motor Rheostat. These high-intensity arcs are 
provided with a small control rheostat in the lamp mechanism motor 
circuit. By manipulation of the rheostat the positive and negative 
carbon feeds may be accelerated or retarded. For most operation 
the arrow on the rheostat knob should point up. Clockwise move- 
ment makes the feed faster, counterclockwise movement slows the 

In making the adjustment the rheostat should be set in closest con- 



FIG. 1. Showing trade terms applying to the high- 
intensity arc. 

f ormity with the burning rate of the positive carbon and any necessary 
manual adjustment should be made with the negative feed control. 
After the feed motor is set to the burning rate of the positive carbon 
the automatic feature previously described, plus the negative-positive 
burning rates of the carbons, will tend to keep the arc in proper posi- 

The various tilt angles at which the lamps are placed affect the 
burning rates of the carbons, but once the lamp is adjusted for a 
given tilt angle the carbon feed control will require very little atten- 

As a means of illustrating proper trimming of the studio types of 
high-intensity carbon arcs the carbon trim for the Type 170 unit is 
shown in Fig. 2. Table 1 shows comparable data for other types. 



Showing Data Regarding the Trimming of M-R Hi- Arcs 



Positive Carbon 

Negative Carbon 





16 mm X 20 in. 

l / 2 in. or 16 /32in. 

IVie in. 

V* in. 

M.P. Studio 

X 8 l / 2 in. M. 

P. Studio 



13.6mm X 22 in. 

Vie in. X 8Vi 

1 6 A in. 


M.P. Studio 

in. M.P. 




9 mm X 20 in. 

7 mm X 9 in. 

13 /ie in. 


High Low Pro- 

Orotip C 



Showing Effect of Varying the Protrusion 

Protrusion (In.) Current Arc Volts Light 

115 76 46 


I 1 /* 130 72 58 

l 3 /i6 138 69 76 

PA 143 68 83 

Operation l/i 148 67 100 


1A 152 64 110 

IVie 165 62 105 

I 1 /* 167 59 97 

1 6 A 168 58 76 

Note. An M-R Type 170 Hi- Arc was used on 115 line volts. The negative 
was kept in the normal position (i. e., that which it assumed with a normal 1 /2-in. 
gap and P/ia-in. protrusion). The positive carbon was successively moved and 
allowed to burn 3 min in each position. 


. Showing Effect of Varying the Gap 

Gap (In.) Current Arc Volts Light 

155* 61 77 


160 63 110 

Normal Operation */2 148 67 100 

Gap Increased ' '" 

* Arc unstable. 

Note. In this test a Type ' 1 70 Hi- Arc was operated on a line voltage of 115. 
in level position and the positive protrusion was kept to the normal ! 5 /i 6 in. The 
gap was varied and the lamp permitted to burn for 3 min in each position. Read- 
ings were taken at the conclusion of each 3-min period. 



Vol 45, No. 4 

The arc will operate satisfactorily if the protrusion is set so the 
lower lip of the positive carbon intersects the axis of the negative 
carbon (Fig. 2). When the lamp is in operation there should be a 
slight under flame lapping the outside of the lower crater wall- as 
shown in Fig. 3. 

Protrusion. Table 2 shows what happens when the protrusion 
of the positive carbon in a Type 170 arc lamp is allowed to vary from 
the normal. Decreasing the protrusion lowers the current and de- 
creases the light. Increasing the protrusion Vie in. raises the light 

16mm X 20" 

I/" np 
/I UK 



FIG. 2. Showing the "normal' 'trimming position 
of the carbons in an M-R Type 170 Hi- Arc. 

level from a normal of 100 to 110 and raises the current from 148 to 
152 amp. Each further increase of protrusion cause's the light level 
to drop and the current to rise. These figures show the advantage of 
keeping the carbons in good trim position. 

Fig. 3 illustrates the arc in proper burning position. Fig. 4 shows 
excessive protrusion and Fig. 5 too short a protrusion. 

Varying Arc Gap. Inasmuch as the arc stream is a path of elec- 
trical resistance the effect of shortening the arc gap is to decrease the 
arc voltage and increase the current and vice versa. Table 3 shows 
the effects of varying the arc gap. 

Carbon Alignment. To obtain maximum performance from a 
studio-type high-intensity lamp the axis of the negative carbon 

Oct., 1945 



must center transversely (in a side-to-side manner) upon the axis of 
the positive carbon. Misalignment of the arc is indicated by a 
tendency of the tail flame to shoot out of the arc at an angle toward 
the right or left. The correction of faulty alignment is a simple 
maintenance problem. 

Varying Line Voltage. Variations of line voltage from stage to 
stage and from lamp to lamp are inevitable in the daily operations 


"' /l ' POSITIVE 



FIG. 3. Showing flame conditions 
of a high-intensity arc in an M-R Type 
170 Hi- Arc. 

of motion picture studios. The effects of line voltage variations 
are shown in Table 4. 

High Line Voltage. In the distribution of current on a motion 
picture set it is not unusual for lamps close to the main entrance 
switch to be operating at high voltage. This is possible particularly 
on large sets where stage voltage may be boosted to overcome the 
voltage drops in long distribution lines. High voltage evidences it- 
self in the negative carbon by a tendency to spindle, that is, to burn 
to a long slender tip. Also under high voltage conditions arc action 
may become erratic and lamp grids tend to heat abnormally. The 



Vol 45, No. 4 

correction within the control of the operator is to lengthen the arc by 
maintaining normal protrusion and increasing the arc gap. Length- 
ening the gap increases the arc resistance and reduces the current. 

Low Line Voltage. In the rigging of sets and in changes of lamp 
position on a set certain circuits may be overloaded with a result 
that lamps in such circuits will be operating at considerably less 
than rated voltage. When high-intensity arcs are operated at lower 

(a) TIP OF 


FIG. 4. Showing the effect of excessive pro- 

than normal voltage the current is reduced, the negative carbon 
burns with a very blunt tip, the arc stream tends to wander around 
the negative carbon tip, and the arc becomes unsteady. A wandering 
arc, a blunt negative carbon tip, and low light output are visible evi- 
dences of low line voltage. 

The remedy within the control of the lamp operator is to shorten 
the gap and keep the arc protrusion close to normal. This will reduce 
the resistance of the arc, bring the current more closely to operating 
standard, and will raise the light level to a marked degree. 

Oct., 1945 



Reverse Polarity. In operation it is essential that the positive 
carbon be connected to the positive side of the line and the negative 
carbon to the negative side. When they are connected in the 
opposite way they are said to be in reverse polarity. 

When a lamp connected in reverse polarity is switched on and 
struck, the arc immediately starts to act erratically. If allowed to 

'/ T.POF 



FIG. 5. Showing the effect of too short a pro- 

burn in this condition for more than a few seconds, the core will be 
burned out of the negative carbon to a considerable depth and fre- 
quently the tip of the positive carbon will be split or shattered. 

When operating a lamp for the first time after it has been connected 
to the line the polarity should be checked by starting the unit. If 
the arc action indicates an error in connection, it should be shut off 
immediately and the pin connectors at the lamp should be reversed. 
Polarity should be checked on all lamps as soon as possible after they 
are connected to avoid possible delays when photographic operations 



Showing Effect of Varying the Line Voltage 

Line Volts Current Arc Volts Light 

f 100 128 58 55 
LmeVoUs ' 

Decreased j n() ^ Q3 gg 

Normal Operation 115 148 67 100 

Line Volts f 120 159 69 118 

Increased \ 125 167* 72 130 

* Arc overloaded. 

Note. In this test a Type 170 Hi- Arc was operated with its trim at all times in 

normal relation (i. e. t with ! 6 /i 6 -in. protrusion and y 2 -in. gap). The voltage was 
varied and readings were taken at the conclusion of a 3-min burning period at 
each successive voltage. 


Filament lamp sources require a minimum of attention on the 
part of studio lighting technicians. Because of this characteristic, 
filament lamp equipments may not get the little attention they should 
have to insure that they give maximum performance and operate at 
proper color temperature. The latter is of particular importance in 
color photography as the spectral quality of the illumination pro- 
vided must be closely controlled. The illustrations and discussions 
which follow are devoted to the larger sizes of filament lamps, i. e., 
the 1-, 2-, and 5-kw bulbs most commonly employed on studio sets. 
Many of the points apply equally to the smaller sizes of lamps. 

Effect of Line Voltage Variations. All incandescent lamp bulbs 
designed for operation in multiple on either d-c or a-c circuits are 
marked as to wattage and voltage. In the case of the 1- and 2-kw 
lamps, the additional marking MP is used to identify the lamps for 
use in black-and-white photography and CP for color photography. 
The 5-kw and 10-kw lamps are suitable for either black-and-white or 
color photography and are marked MP-CP which indicates that they 
are interchangeable. In the future, the MP-CP marking may be 
deleted as their application in airport floodlighting is increasing. 

Fig. 6 shows the effect of varying line voltage on the light output, 
watts consumed, and color temperature of the CP line of studio lamps. 
The curves apply to lamps marked 120 v and the 3 curves cross at the 
120-v point on the chart. The curves would be identical in shape and 
slope if drawn for lamps marked 115 v; however, the curves would 

Oct., 1945 



then cross the 115-v point on the chart at a value of 3350 K and 100 
per cent initial light output and watts. This simply means that lamps 
labeled 120 v must have 120 v at the lamp terminals (not at the panel 
board) if rated light output is to be obtained, normal wattage con- 
sumed, and proper color temperature realized. Similarly, lamps 
labeled 115 v must be operated at 115 v to attain designed performance. 
If the lamp socket voltage is 5 v below rated lamp voltage, the watts 





, 80 
O 70 




O 40 


5 30 






*/ATTS fa 



OR TEfv 






1-2-5 H 











1 ^ 




> 120 125 

FIG. 6. Some characteristics of gas-filled tungsten lamps when 
operated at voltages other than at their rated design values. Ac- 
curate voltage control is of particular importance in color pho- 

will be reduced approximately 6 per cent, color temperature about 55 
K, and light output 14 per cent. 

Unfortunately, the human eye cannot readily detect light output 
and color temperature variations caused by a 5- or 10-v drop in feeder 
voltage. A voltmeter, connected across the bulb terminals in the 
lamp house, is the most positive check and one that can be made 

Low socket voltage is most often attributed to overloaded feeder 
cable. However, other factors, such as burned switch points and 
poor connections at plugging boxes and at the lamp socket itself 



Vol 45, No. 4 

can be contributing factors. A 5-kw incandescent bulb operates at 
approximately 45 amp. Poor connections at the bipost lamp ter- 
minals at this current value will cause rapid oxidation and pitting of 
the lamp posts and socket sleeves. These connectors require fre- 
quent inspection to insure clean and tight contacts. Bipost sockets 
designed to grip firmly the entire lamp post rather than to cut into the 
copper shell at one point with a set screw, are preferred types. A 
poor lamp connection can be quickly checked by removing the bulb 
from the holder and inspecting the lamp posts for oxidation pits. 

10 20 30 40 50 60 70 80 90 


FIG. 7. Total light output of high-wattage lamps can be restored to ap- 
proximately their initial value by swirling the tungsten powder in the bulb. 
Although the total light output in lamp bulbs not cleaned falls off as shown 
above, most of the blackening occurs on the glass bulb directly above the fila- 
ment and the useful light passed through lens-type spotlamps is not reduced 
in these amounts. 

Importance of Tungsten Cleaning Powder. High-wattage lamp 
bulbs in the 2-, 5-, and 10-kw sizes are available with or without 
tungsten cleaning powder. Proper use of this cleaning agent, Fig. 
7, results in lamp performance at close to initial light output through- 
out the lamp's useful life. Good maintenance practice indicates that 
the bulb should be removed from the lamp house after every 20 hr, 
and preferably every 10 hr, of burning and the tungsten powder 
swirled about in the bulb. It is appreciated that accurate records of 
burning hours cannot be kept on studio lamps. However, whenever 
lamp bulbs are removed from the units for any reason or when an 

Oct., 1945 



inspection reveals that considerable blackening has accumulated, it is 
worth the effort to shake the powder about the bulb for a few seconds. 
Although some technicians prefer not to remove bulbs until they have 
burned out, because of the hazard of dropping or breaking the bulb, 
Fig. 7 indicates that worth-while gains in total light output are 
possible if a reasonable amount of care is exercised when cleaning 

In addition to the gain in over-all light output throughout life 
when lamps are cleaned, longer useful life results as the lamp bulb 
operates cooler and has less tendency to blister when the blackening is 
periodically removed. This is particularly true in the case of -the 

K . 



5KW T-64(G-64) LAMP 






70 75 

FIG. 8. Color temperature characteristic of 5-kw studio lamps 
during life applicable to lamps that are cleaned at least every 20 hr 
of burning by swirling the tungsten cleaning powder about in the 

spherical or G64 bulb lamps as their bulb volume is approximately 25 
per cent less, and the heat correspondingly greater than in the im- 
proved tubular T48 bulb lamps. 

Color Temperature during Lamp Life. In the past, color tem- 
perature of incandescent lamp bulbs was rated in terms of initial 
performance. The CP line of lamps had an initial rating of 3380 K. 
A more useful rating is the mean color temperature value, and all 
lamps used for photographic purposes are so rated today. This 
change is in rating only as no changes in physical or electrical charac- 
teristics have been made. Fig. 8 shows the color temperature per- 
formance throughout life of 5-kw lamps which have an average rated 
life of approximately 75 hr. Midway in life, the color temperature of 
the average lamp is approximately 3350 K, which is approximately 30 
degrees lower than it was initially and will drop approximately 20 


degrees more when the lamp reaches rated life. If the color require- 
ments on a particular set are such that use of the lamp should not be 
continued, even though it is still operative after some 75 hr of use, 
it can be used on black-and-white sets in either lens spotlamps, sun- 
spots, or skypans. The color temperature at this point in life is still 
appreciably above that in the lower wattage line of MP lamps, which 
are designed for certain life values rather than for a mean color 
temperature value. 

General Maintenance Hints. All of the above discussion has 
been directed to the operation and maintenance of the lamp bulb 
itself. Other parts of the lighting equipments require periodic main- 
tenance if maximum over-all efficiency of the generated light is to be 
utilized. Dirt and film on glass and metal reflectors absorb con- 
siderable light if allowed to go any appreciable time without cleaning. 
It should be remembered that light from the lamp bulb must go 
through any layer of dirt on the mirror twice once, in getting to the 
specular surface, and again when leaving it. Special compounds are 
available for quickly cleaning without injury to the specular surfaces. 
Similarly, condenser lenses, such as the Fresnel type, must be kept 
clean for maximum efficiency. 

Studio Lighting Committee 

C. W. HANDLEY, Chairman 




1 LINDERMAN, R. G., HANDLEY, C. W., AND RooGERS, A.: "Illumination in 
Motion Picture Production," /. Soc. Mot. PicL Eng., XL (June, 1943), p. 333. 



The Committee on Standards of the Society of Motion Picture 
Engineers has been active in recent months primarily through in- 
dividual memberships in the various subcommittees of War Stand- 
ards Committee on Photography and Cinematography, Z52, of the 
American Standards Association, where the urgent needs of the Armed 
Services have resulted in an unprecedented rate of motion picture 
standardization, particularly in the 16-mm field. These War Stand- 
ards are now under consideration from a peacetime point of view, with 
the early objective of establishing those suited to the peacetime 
economy on a more permanent basis. 

The Committee has under consideration a total of 8 other proj- 
ects, of which only 2 have been receiving attention during wartime. 
The first of these has to do with the dimensional characteristics of 
35-mm projector sprockets where a slight increase in diameter gives 
promise of providing a substantial increase in film life. Dr. Carver 
expects to have a report 1 of his subcommittee's activities in this 
connection ready for presentation at the Fifty-Seventh Semi- Annual 
Technical Conference. A second project is the development of a 
glossary of terms used in the motion picture industry, which has not 
been pushed very hard because of more important wartime demands. 
Nevertheless, a considerable amount of progress has been made and 
a more active prosecution of this project is being planned for the 
early future. 

In anticipation of a more aggressive post-war program, the mem- 
bership of the Committee on Standards has recently been enlarged 
to 47 individuals, thus providing a more representative group in- 
cluding all fields of the motion picture industry. Many of the new 
members were chosen because, along with older members, their ex- 
cellent contributions to the activities of the ASA War Committee 
have indicated unusual ability in their particular field. Thus, al- 
though the Committee on Standards as an organization has been in 
comparative hibernation during the war years, the membership is 
being maintained in a state of active training for the peacetime work 
to come. 

F. T. BOWDITCH, Chairman 

* Presented May 14, 1945, at the Technical Conference in Hollywood. 

1 "Report of the Subcommittee on Projector Sprocket Design," /. Soc. Mot. 
Pict. Eng., 45, 2 (Aug., 1945), p. 73. 



The basic assignment of this Subcommittee of the Theater Engi- 
neering Committee of the Society has been, and still is, to find, or 
cause to be developed, instruments for measuring the uniformity and 
level of screen brightness and screen illumination. Since no instru- 
ment has been available having properties desired by the Subcommit- 
tee as outlined in a previous report, 1 development work on the part of 
instrument manufacturers is needed. Such development work has 
been out of the question during the war and the Subcommittee has 
therefore been unable to proceed. 

In anticipation of relaxed pressure on development engineers in 
connection with war projects, the Subcommittee is now again active. 
The original specifications have been reviewed and approved, and 
letters have been sent to instrument manufacturers inviting their 
consideration of this problem. It is still too soon to know which ones 
may be willing to docket this project as very few replies have been 
received so far. 

F. E. CARLSON, Chairman 


1 "Report of Subcommittee on Screen Brightness," /. Soc. Mot. Pict. Eng., 
XXXVIII, 1 (Jan., 1942), p. 81. 


There has recently been formed a new committee of the Society, 
the Subcommittee on Theater Engineering, Construction, and Opera- 
tion. This is a subcommittee of the Theater Engineering Commit- 
tee which is under the chairmanship of Dr. Alfred N. Goldsmith. 
The membership is as follows : 








* Presented May 14, 1945 at the Technical Conference in Hollywood. 


We believe that this Subcommittee will be most effective in ex- 
tending the activities of the Society by offering needed technical 
assistance to the motion picture theater industry. The Subcom- 
mittee intends to function in 2 principal ways : 

First, it plans to investigate a number of special subjects relating 
to theaters, and to develop information valuable to the theater opera- 
tor on such subjects as Theater Illumination, Ventilation, Cooling, 
and Air Conditioning, Acoustics, Seating, Theater Carpets, etc. 

The Subcommittee will make a special effort to discover new ap- 
plications to theater construction and operation of materials, proc- 
esses, and inventions which have arisen out of the war effort. 

The Subcommittee has already instituted a study of the subject 
of Theater Carpets. The whole group of subjects is highly technical 
and worthy of study by the Society's best engineering talent. It has 
been learned that while carpets have been developed to a high de- 
gree of perfection over a period of several thousand years by trial 
and error methods, there is nevertheless much research work being 
carried on by the larger manufacturers, and the Association of Carpet 
Manufacturers has an active Research Committee. 

The manufacturers have been discouraged to find that, in spite 
of the efforts devoted to improve their product, a great deal of the 
effort is nullified by the consumer, of which the theater is one, who 
may in general be said to install and maintain his carpets in an un- 
scientific, even unintelligent, manner. The Subcommittee is con- 
vinced, as are also the manufacturers, that in view of the huge sums 
spent by the theater industry for carpets, it may be instrumental 
in saving the motion picture industry very substantial sums of money 
now wasted because of improper carpet specification, installation, and 

Secondly, the Subcommittee will establish a suggested set of stand- 
ards for theater construction and operation. At the present time 
numerous cities and states are extensively revising their building 
codes, and such codes in all instances contain special sections relat- 
ing to theaters. Most of these codes have been developed by carry- 
ing along old untried specifications plus the introduction of new 
equally untried specifications. The industry has been at a dis- 
advantage in meeting with code makers, for the industry has been 
poorly equipped with technical information necessary for dependable 
and useful standardization. 

It is the intent of this Subcommittee to assist the industry in the 


development of reasonable codes, by endeavoring to develop reason- 
able standards for such things as theater seating, theater illumination, 
stage curtains, stage skylights, exit facilities, and many other factors, 
that go into the development of a code. 

The Chairman of the Subcommittee is a member of a committee 
appointed by the State of New York Labor Department which is 
now formulating a code covering all places of public assembly in the 
State. He is, therefore, informed upon the needs of a code-making 
authority. He has reported that the Recommended Standards of the 
Society of Motion Picture Engineers for Motion Picture Projection 
Rooms have been of great assistance to the New York State Com- 
mittee and that the Society's standards will to a large degree be in- 
cluded in the new State code. 

The Subcommittee will coordinate its efforts with the other com- 
mittees of the Society. It will value comments and suggestions 
from the membership. 





Summary. Eastman Fine-Grain Sound Recording Film Types 1372 and 
1373 have been introduced recently to the motion picture trade. Type 1372 is a film 
having high contrast and with capabilities of producing a high degree of image sharp* 
ness for variable-area recording. This emulsion may be used for recording with 
ultraviolet or white light with very little difference in sound quality. High energy 
developers are not necessary for the development of this film owing to the inherent lack 
of image spread. Type 1373 is a film designed for development in a normal picture 
developer for variable-density recording. The use of special low energy developers is 
therefore avoided. Test data are presented for these 2 films. 

The development of suitable sound films has entailed the meeting 
of a great number of requirements and there have always been some 
that were in direct conflict. There have been many cases where a 
film that was capable of giving very good results did not fit into the 
methods of operation and was therefore judged unsatisfactory. It is 
required that a sound negative material be capable of being exposed 
on the type of recorder in use and so developed that when printed 
upon the release print and developed with it under conditions that 
give the best in picture quality, the sound will also be the best obtain- 
able. Changes in recording technique and in materials used in mak- 
ing prints may call for a new sound film. In developing these 2 new 
Eastman Sound Films, Type 1372 for variable-area recording and 
Type 1373 for variable-density recording, all latest developments 
have been taken into account. Where conflicting requirements have 
been encountered, the requirement of sound quality has been favored. 


Prior to 1932, the Eastman film which was used for sound recording 
purposes was Eastman Release Positive, Type 1301. This film was, 

* Presented October 17, 1944, at the Technical Conference in New York. 
** Motion Picture Film Dept., Eastman Kodak Co., Rochester, N. Y. 

f Motion Picture Film Dept., Eastman Kodak Co., Hollywood, 
tf Motion Picture Film Dept., Eastman Kodak Co., New York. 



of course, designed for use as a print stock but was found to give 
reasonably satisfactory results when used for either variable-density 
or variable-area sound records. At that time, however, the emulsion 
speed was slightly low for variable-density recording. Also, it was a 
high-gamma material and development to the low gamma required 
made control of development rather difficult. 

In 1932, Eastman Sound Recording Film, Type 1359, was made 
available for variable-density recording. This film had considerably 
more emulsion speed than 1301. It was, like 1301, however, a high 
contrast film. In 1936, for variable-area recording with ultraviolet 
light, Eastman Sound Recording Film, Type 1357, was submitted to 
the trade. Gradually Type 1357 supplanted Type 1359 for use as a 
variable-density recording medium. 

In 1937, the first Eastman Fine-Grain Sound Recording Film was 
received in Hollywood. This film, Type 1360, was orthochromatic 
and was designed for use in variable-area recording. It was found to 
be superior to Type 1357 because of improved high-frequency re- 
sponse and reduced background noise. Type 1360 was also found 
to give improved sound quality over the older materials when used as 
a rerecording print stock for rerecording purposes. It had the dis- 
advantage of being finer grain and having much less image spread 
than the print stock then in use, and a very high density was required 
to produce satisfactory cross-modulation. This made the effective 
speed too low for many users. 

About 1937 there was generated a great amount of interest in the 
use of fine-grain films for sound, and experiments were run using all 
the available fine-grain films for recording and rerecording. In 1939, 
Type 1361 was introduced as a fine-grain rerecording print stock with 
the idea that a fine-grain material at this point would serve as much 
purpose as a fine-grain negative. Later in 1939, however, Type 
1302, Eastman Fine-Grain Release Positive, was introduced and 
there was no further need for a film like Type 1361. 

Up to that time there was no fine-grain sound recording film avail- 
able with a satisfactorily slow developing rate for variable- density 
negatives. In fact, all the films used for variable-density sound 
recording up to that time, whether fine grain or non-fine grain, had 
been of the inherently high-gamma infinity type. 

In November, 1939, Eastman Fine-Grain Sound Recording Film, 
Type 1366, which possessed a slow development rate, was made avail- 
able. This film was used successfully at low gamma values for 


variable-density sound recording. A very low control gamma was 
required for this film and under that condition the speed was not 
quite great enough for most users. 

Since all fine-grain films which had been manufactured up to 1939 
had given brown-toned screen images, because of the inherent light- 
scattering properties of very fine silver bromide grains, no great in- 
terest had been shown in using these films for release print stock. A 
method was found for producing a black-toned image on Eastman 
Fine-Grain Release Positive, Type 1302, late in 1939, and this film 
has found widespread use as a release print material since that time. 
It is used to a limited extent for variable-area recording with white 
light and serves very well for this purpose. It is not deemed satis- 
factory as a variable-density sound recording stock because of its 
inherent high gamma and rapid development rate. 

Fine-Grain Sound Recording Film, Code 1370, which was first sold 
in 1940, was a low gamma infinity emulsion capable of high resolving 
power. However, it still required the use of the so-called variable - 
density sound type of developer, which produces a slower developing 
rate than normal picture negative developer. This type of developer 
has always been necessary for development of variable-density sound 
track, because of the necessity for obtaining extremely low gamma 
values on films which have generally been of the high gamma infinity, 
or positive-film type. 

The need was apparent for a variable-density sound recording 
film, having low gamma infinity, which could be handled at normal 
times of development in a regular picture negative developer. Ex- 
periments were started toward that end, and early in 1944 Eastman 
Fine-Grain Sound Recording Film, Type 1373, was produced. 

The rapid growth of 16-mm production during the past few years 
had emphasized the need for a variable-area sound recording film 
capable of extremely high resolution. To satisfy this demand, 
Eastman Fine-Grain Sound Recording Film, Type 5372, in 16-mm 
width, was produced in 1942. This film has received wide accept- 
ance by the 16-mm professional trade and has proved to be a superior 
product in all ways. Extensive tests were begun, with the coopera- 
tion of RCA on the same emulsion in 35-mm width, which have 
shown that this film is superior for 35-mm variable-area sound re- 

At the present time, the Eastman films which are being used by 
the motion picture industry for variable-area sound recording are 



Vol 45, No. 4 

Types 1357, 1302, and 1372. Those being used for variable-density 
sound recording are Types 1357, 1301, and 1373. In the considera- 
tion of the emulsion characteristics, the physical characteristics, and 
the results of sound quality tests on 1372 and 1373, comparisons will 
be made with other Eastman films now being used for sound recording 

FIG. 1. Eastman Sound Recording films, exposed to tung- 
sten and to ultraviolet radiation, lib sensitometer. Average 
Hollywood conditions of use. 


Advantages. Type 1372 emulsion possesses several outstanding 
emulsion characteristics when compared with older well-known 
variable-area sound recording films, such as Types 1302 and 1357. 
These are enumerated as follows : 

(1) Emulsion speed comparable to 1357 when both are exposed with ultraviolet 
light; higher speed if 1372 is exposed with white light; and 1357 to ultraviolet 

(2) Higher resolving power and less image spread. 

(5) Emulsion sensitivity confined largely to that region of wavelengths to which 
the emulsion shows strong absorption, i. e., the ultraviolet and violet region. Thus 
there is little need for ultraviolet filters in exposing this film. 


(4) Sufficient image sharpness for use as a sound negative when developed in 
ordinary positive developers of the D-16 type; does not necessitate the use of 
special high-energy developers now used for all 35-mm variable-area sound re- 
cording films, 

Emulsion Characteristics. In Fig. 1 are given typical H and D 
curves for the Eastman sound recording films as they are used at the 
present time. The ultraviolet exposure curves for both 1357 and 
1372 were made by replacing the standard positive conversion filter 

EMULSION I372-0: 1 




- 2 


DENSirr or BASE 


FIG. 2. Sensitometric curves of Eastman Fine-Grain Sound Re- 
cording Film, Type 1372. Typical Hollywood processing conditions. 

(No. 78B) in the lib sensitometer with a Corning 584 filter, 0.75 mm 
thick. The log R scale given in Fig. 1 is, therefore, not a true scale 
for the ultraviolet exposures. 

All ultraviolet recorder exposures mentioned in this paper were 
also made with the 0.75-mm Corning 584 filter. 1 The Corning 597 
filter, 3 mm thick, which is also used for ultraviolet recording, has 
considerably wider transmission limits, and thus produces more 
exposure on either 1357 or 1372 emulsions. 

The high-contrast curves on 1372, 1357, and 1302 in Fig. 1 are the 
result of 7 min development in the variable-area sound negative de- 
veloper in a representative laboratory. This constant time of de- 



Vol 45, No. 4 

velopment for all films is fairly typical of trade practice. At such 
development conditions, the films used for variable-area sound re- 
cording have reached, or at any rate closely approached, gamma 
infinity for the type of developer used. In comparing 1372 with 
1357, the considerably higher contrast of 1372 is noteworthy; in 
comparing 1372 with 1302 the large advantage in emulsion speed of 
1372 when both films are exposed with white light is evident. The 
large differences in the "toe shapes" of these 3 emulsions are quite 
apparent from the typical curves given in Fig. 1. 

EMULSION 1372-02 



2.0 - 


FIG. 3. Sensitometric curves of Eastman Fine-Grain Sound Re- 
cording Film, Type 1372. Developed in D-16 under same conditions 
of agitation used for curves obtained in Fig. 2. 

For variable-area recording a high contrast is desirable so that the 
exposed areas will develop to a high density while the unexposed 
areas remain clear; the boundary between the two must be as sharp 
as possible. Thus, the ability of the 1372 emulsion to reproduce 
geometrically sharp edges, which is known as sharpness, is a factor 
of prime importance in the choice of an emulsion for variable-area 
sound recording. Since the density gradient at the edge of the image 
of a geometrically sharp-edged object in general follows the same 
curve as does the H and D characteristic of the emulsion, it is appar- 


ent that not only gamma but also toe shape is of importance in the 
selection of an emulsion to give maximum resolution of high fre- 
quencies in a variable-area sound record. 2 The resolving power of 
1372 is 150 lines per mm, 1302 resolves 90 lines per mm, and 1357 
resolves 50 lines per mm. The manner of arriving at such test values 
of resolving power has been discussed by Mees. 3 

Figs. 2 and 3 give families of sensitometric curves obtained with 
1372, when this film is developed in a typical Hollywood variable-area 
sound negative developer and in D-16. Both developments were 
carried out in the same developing machine, and the rates of de- 
velopment are believed to be fairly representative of laboratory 

There are 3 developer formulas which should be considered in 
a study of development conditions for 1372. These are given in 
Table 1. 


Developers Used for Variable- Area Sound Recording Films 

Release Positive V.-A. Sound Negative 
Chemical D-16 Developer Developer 

Elon 0.3 1.5 2.0 

Hydroquinone 6.0 3.0 11.0 

Sodium sulfite, des. 38 . 40 . 40 . 

Potassium bromide 0.9 2.0 2.0 

Sodium carbonate, des. 19.0 ? ? 

Citric acid 0.7 

Potassium metabisulfite 1.4 

pR 10.0 10.0 10.4 

All quantities expressed as grams per liter of solution. 

The formula entitled "Variable- Area Sound Negative Developer" 
is representative of developers of that type used by motion picture 
laboratories, but is not to be regarded as the best recommended form- 
ula for use under all sets of operating conditions. This formula was 
used to obtain the curves shown in Fig. 2. The formula entitled 
"Release Positive Developer" is believed to be representative of 
positive developers now in use for developing motion picture film. 
It should be noted, however, that there is wide variation in the 
chemical composition of -positive developers used by motion picture 
laboratories, so that statements made with reference to the formula 
given in Table 1 should not be construed as applying to all positive 


It has been found that development of 1372 in the "Release Posi- 
tive Developer" of Table 1 does not give as satisfactory results as does 
development in D-16 or the "Variable- Area Sound Negative De- 
veloper." Gamma values are much lower, toe break density values 
are higher, and cancellation tests show the tolerance of operating 
limits to be narrower than those obtained in either of the other 2 
formulas. It will be noted that the hydroquinone content of the 
"Release Positive Developer" is low in comparison with that of the 
other developers. It is believed that tests should be made in the 
regular positive developer at any laboratory at which emulsion 1 372 
is to be developed before the decision is made as to whether a satis- 
factory variable-area sound negative can be obtained in this manner. 
Formulas of the D-16 type have been used quite successfully in some 
commercial laboratories for the development of 16-mm 5372 as well as 
either regular or fine-grain 16-mm release print stocks. 

At the present time 4 Hollywood laboratories maintain special high- 
energy developers for the development of variable -area sound nega- 
tive and rerecording prints. One Hollywood laboratory has used 
for this purpose the regular positive developer at prolonged times of 
development. The other Hollywood laboratories do not handle any 
variable-area sound track. To date all production work on 35-mm 
1372 done in Hollywood has been developed in the high-energy type 
of developer, but from a study of the results in D-16 one is led to the 
conclusion that where it is possible to use a positive developer with a 
higher developing agent concentration and a higher hydroquinone- 
elon ratio, Type 1372 can be handled in the same developer and thus 
eliminate the necessity of a special sound developer. 

From a study of the families of curves in Figs. 2 and 3, it may be 
seen that there are a multiplicity of combinations of exposure and 
development conditions which lead to equivalent densities at very 
nearly equivalent gamma values. While, on the other hand, this 
fact allows considerable leeway in the exposure and development 
conditions, yet, on the other hand, it behooves the sound and labora- 
tory technicians to determine, by a series of controlled tests, which 
combination of these conditions results in the best quality. 

As stated earlier in the paper, it is not necessary to use an ultra- 
violet filter with Type 1372 since the image spread is very low. Since 
a filter is not needed the effective speed is high, being about 2 l /z times 
that of Type 1357 with the commonly used ultraviolet filters. If, 
however, it is thought desirable to use a filter, the spectral sensitivity 


of Type 1372 is such that there is a minimum speed loss with the in- 
troduction of a violet filter. With the commonly used filters, Type 
1372 is approximately as fast as Type 1357 with the same filter. 

Fig. 4, the wedge spectrogram of 1372 emulsion, shows that the 
sensitivity of this film extends to about 460 m/z at the long wave- 
length end. This value is somewhat lower than the corresponding 
long wavelength limit of sensitivity of other sound recording emul- 
sions. This fact, coupled with the relatively high emulsion speed, 
evident in Fig. 1, is proof that 1372 possesses a much higher ultra- 
violet to white light sensitivity ratio than do other sound recording 
films. The measured ratios of lib ultraviolet to white light sensi- 
tivities are: 13571 to 10; 13021 to 8; 1372 I to 3. These 
values represent the ratio of exposure necessary to produce densities 
of 2.0 for equal development times. 




FIG. 4. Spectrogram of 1372 to tungsten light. 

Under the conditions of forced development used for variable-area 
sound records, safelight fogging becomes a matter of real concern. 
Too frequently, it is believed, fog which is attributed to the chemical 
action of the high-potential developer, is actually the result of too 
long exposure to light of very low intensity level. Results of an 
exaggerated safelight fogging test on 1372 and other films used for 
sound recording are given in Table 2. Test conditions were arbi- 
trarily chosen. An OA safelight, 8 X 10 in. in size, containing a 15-w 
lamp, was placed at a distance of 3 ft from the film emulsion surface, 
and the direct radiation from the safelight allowed to fall on the film. 
Development was for normal time in a variable-area sound negative 


Safelight Fog Densities 

Exposure Time to OA Safelight, in Min 
Emulsion 2 5 10 

1372 0.02 0.02 0.02 0.02 
1302 0.02 0.02 0.04 0.06 
1357 0.04 0.06 0.08 * 0.16 



Vol 45, No. 4 

The support upon which 1372 emulsion is coated carries a gray 
anti-halation backing which serves to reduce still further the many 
possible causes of an unsharp image. 

Type 1372 has a thinner emulsion than either 1357 or 1302 and re- 
corders should undoubtedly be refocused for optimum results on this 
film. Those made with and without focusing, however, have not 
shown too great a difference. 

Sound Quality Test Results. Examination of high-intensity 
H and D curves obtained on 1372, using both ultraviolet and white 

light, shows the curve shapes to 
be identical with lib curves 
which have been given the same 
development. The high-inten- 
sity exposures were made on an 
improvised sound recorder. The 
light intensity was modulated 
with a series of neutral density 
filters, and the exposure time 
approximated the mean value 
used in an actual sound record- 

Results of recorder exposure 
tests, high-frequency attenuation 
measurements, and cross-modu- 
lation tests made by RCA on 
Type 1372 have been reported in 
another paper presented before 
this Society. 4 Consequently, 
only a brief summation of the 


1.30-8 - 


110-16 - 

200 220 240 260 280 3.00 

Fig. 5. Cross-modulation tests, nega- 
tive developed in D-16. 

advantages of this film from the standpoint of sound quality will be 
presented here. 

Listening tests give emulsion 1372 a positive advantage in quality 
over the older types of film, particularly because of the absence of 
background noise of the type which is usually considered to be in- 
herent in the film, i. e., caused by graininess, fog, and film surface 
characteristics . 

Cross-modulation tests, developed in either D-16 (Fig. 5) or the 
special variable-area sound negative developers now used in Holly- 
wood, show 1372 to have wider print density tolerance limits for 
30 db cancellation than any of the older types of film. When used 


in place of 1302 for a rerecording print stock, the tolerance limits are 
still broader. The optimum negative density is in the range 2.6 to 
2.8 for best cancellation at a normal print density of about 1.2 to 1.4, 
when the negative is developed to the lib gamma control value of 
approximately 3.0 to 3.5. These limits apply to either ultraviolet or 
white light exposure of the 1372 negative. For these negative condi- 
tions, an optimum rerecording print density of 1.8 to 2.0 is found for 
1302 and 2.5 to 2.7 for 1372. 

The high-frequency response is improved to the extent of about 2 
db at 9000 cycles when 1372 is compared with 1357 as a recording 
stock and both are printed on 1302. This gain is noted with 1372 
for either a rerecording print or release print development conditions. 
The use of 1372 as a print stock adds a further improvement in high- 
frequency response. 

The comparatively small amount of image spread in 1372 emulsion 
makes it possible to use it as a direct positive at densities of from 1.2 
to 1.3, whereas most films require a density of about 0.6 to 0.8 for mini- 
mum distortion. At this low density level, the volume output is too 
low and the noise level too high for satisfactory use of these films as 
direct positives. 

When 1372 is used for rerecording prints or for daily prints, a loss 
in volume, attributable to the gray base density, of approximately 
2 db is found. Since this volume change is barely audible, it is the 
practice of some studios now using this sound film to intercut prints 
made on 1372 out-take negative film with prints made on 1302. In 
at least one instance, for the sake of uniformity of quality level, the 
sound department has requested the laboratory to purchase 1372 for 
use as a print stock for sound daily and rerecording prints instead of 
the customary 1302. 


Advantages. Just as improvements in sound recording equip- 
ment and technique have been many and rapid within recent years, 
so has it become increasingly evident that films for recording pur- 
poses must be designed to do the best possible job under the par- 
ticular set of conditions which the sound engineer intends to impose 
upon them. This means that the older types of variable- density 
sound recording films, which were designed by the emulsion makers 
originally as either print stocks, or high gamma infinity dual-purpose 
sound recording films, have been found lacking. 


Type 1373 emulsion differs from the older Eastman films which are 
used for variable-density sound recording in 2 primary respects: 
(1) It is finer grained and is thus capable of higher resolution and 
produces less background noise. (2) It is an inherently low gamma 
infinity emulsion, and has a considerably slower developing rate than 
other variable-density sound recording films. 

A number of practical benefits derive from the second primary 
advantage mentioned above. They are : 

(1) Marked reduction in directional effects . 

(2} Elimination of special low-energy developer; 1373 may be developed in 
picture negative developer. 

(3) Use by the laboratory of developing machines having best agitation condi- 
tions for picture negative development, as well as for sound negative develop- 
ment. At present this is not possible in most laboratories. 

(4} Greater latitude in development control, owing to the flat characteristic 
of time-gamma curve. 

Emulsion Characteristics. In Fig. 1 are given comparative 
sensitometric curves for 1357, 1301, and 1373 when these emulsions 
are all developed to a gamma value of 0.55. This value of gamma 
was chosen for all 3 films as being typical of Hollywood practice, 
when variable-density sound negatives are printed onto 1302 with 
ultraviolet light. When this practice is followed, it is found that 
0.55 represents the average control gamma value on these 3 sound 
recording films. If white light printing is used, the required lib con- 
trol gamma value is about 0.35 to 0.40 for all 3 emulsions, and the 
relative positions of the curves in Fig. 1 remain unchanged. 

Until the introduction of Type 1373, all variable-density sound 
recording negative required development in special low-energy de- 
velopers which allowed the development process to take place suffi- 
ciently slowly so that the low gamma values required would be ob- 
tained in from 4 to 8 min. If the same films, such as 1357 or 1301, 
were developed in regular picture negative developer, it was found 
that generally the minimum gamma value obtainable, with the 
developing machine speed set at maximum, was considerably higher 
than even the value of 0.55 required for ultraviolet printing. 5 Fur- 
ther, if the machine thread-up was so altered that the required gamma 
values could be obtained, it was found that the developing time 
under such conditions was of the order of 2 to 3 min, which is un- 
questionably too short for uniform development of a sound recording 


At present at least 6 major Hollywood laboratories maintain such 
low-energy developers, primarily for the purpose of developing 
variable-density sound negatives. Referring again to Fig. 1, the 
curves for 1357 and 1301 were obtained by developing these films 
for 7 and 6 min, respectively, in a typical Hollywood variable-density 
sound negative developer, while the curve for 1373 was obtained by 
developing this film for 6 min in a typical Hollywood picture negative 

Fig. 6 shows a family of sensitometric curves, as well as time- 
gamma and time-fog curves obtained on 1373 in a normal picture nega- 

EMULSION: 1373-03 


AT 68 F 




FIG. 6. Sensitometric curves of Eastman Fine-Grain Sound Recording 
Film, Type 1373. Typical Hollywood processing conditions, picture negative 

tive developer under machine conditions considered to be about 
average in Hollywood. It is of interest to note the relatively small 
variation in gamma which accompanies a large variation in developing 
time. This is of considerable aid to the laboratory man whose prob- 
lem is the maintaining of an exact gamma value from day to day on 
the sound negative stock. It is suggested that 4 times normal ex- 
posure be used in making control strips on the lib sensitometer on 
1373, so as to obtain a full scale of density values. 

In those cases where low gamma values, such as 0.35 to 0.40, are 
required for white light printing, it is apparent that the developing 
time of 1373 in picture negative developer may become impracticably 
short. In some laboratories, where the lower limit of developing 



Vol 45, No. 4 

time in their picture negative machines is too high to permit obtain- 
ing such low gammas on 1373, it would then be necessary to use the 
variable-density sound negative developer. Typical sensitometric 
results obtained in such a developer are shown in Fig. 7. Tests made 
with 1373 emulsion have shown that there is no valid reason for ob- 
jection to the use of the low-energy sound negative developer so far 
as sensitometric considerations are concerned. At equivalent gamma 
values, the densities obtained in the sound negative and picture nega- 
tive developers are very nearly equivalent in several laboratories 
where such tests have been made. Likewise, the curve shapes are 

EMULSION: 1373 - 03 


AT 68 F 


FIG. 7. Sensitometric curves of Eastman Fine-Grain Sound Re- 
cording Film, Type 1373. Typical Hollywood processing condi- 
tions, variable-density sound negative developer. 

identical within limits of experimental error. Formulas of the de- 
velopers used for obtaining the data in Figs. 6 and 7 are given in 
Table 3. 


Developers Used for Variable-Density Sound Recording Films 



Sodium sulfite, des. 
Potassium bromide 

Picture Negative 


V.-D. Sound Negative 


All quantities expressed as grams per liter of solution. 


It is well known that the Eberhard effect, which manifests itself 
in motion picture film development as a "directional effect," is the 
greatest for moderate degrees of development, diminishes when the 
development is prolonged, and becomes negligible as the develop- 
ment approaches the limit. In other words, approach to gamma in- 
finity permits the development of those areas which may have lagged 
because of excessive concentration of reaction products during the 
first and intermediate stages of development to catch up. As pointed 
out by Ives and Jensen, 6 Leshing, Ingman, and Pier, 7 and others, the 
lack of density uniformity because of the streaming of reaction prod- 
ucts from contiguous areas in a variable-density sound track may be 
pronounced even with a reasonable degree of agitation. There is no 
other case in the processing of motion picture films where the develop- 

FIG. 8. 

Directional effect on several variable-density sound recording 
Regular lib sensitometric step-tablet. Dots, 2.4-mm 

diameter, 1 cm apart. 

ment process is stopped so far from completion. The need for a 
variable-density sound recording film with inherently low gamma in- 
finity which could, therefore, be developed so that the Eberhard 
effect is negligible, was obvious. 

Leshing and Ingman 8 have employed the method of comparing 
densities of small dots with densities of large, contiguous areas, both 
having been given the same exposure, in order to study the degree of 
agitation in a developing machine. This same technique lends itself 
admirably to the comparison of various sound recording films under 
the same conditions of development. In Fig. 8 are given the char- 
acteristic curves obtained upon 1357, 1301, and 1373. All these 
films except 1373 are of the inherently high-gamma infinity type. 
Along with normal lib sensitometric curves, which are obtained by 
reading areas of increasing density approximately one centimeter 
square, are given curves obtained from the readings made on small 


circles of density, 2.4 mm in diameter with relatively large unexposed 
areas between steps. It will be noted that the spread between the 
pairs of curves is abnormally large in the case of the high gamma in- 
finity films and very much less for 1373. These data were obtained 
under actual working conditions in a Hollywood laboratory. De- 
velopment time was 6 min for all films, with machine speed and agita- 
tion conditions identical, and the only variable, other than emulsion 
type, was the developer formula. In order that a reasonable match 
in gamma values be obtained, picture negative developer was used for 
the other films. It is believed that the difference in composition of 
the 2 developers used in making this test can account for very little 
of the difference found in directional effect. 

Similar tests to those shown in Fig. 8 were made in other Holly- 
wood laboratories. These tests lead to the following conclusions : 

(1) Under conditions of actual laboratory practice, with differences in machine 
agitation, developing time, and developer formula, emulsion 1373, in all cases, 
showed very much less directional effect than did the other 3 emulsions tested. 

(2) The magnitude of the directional effect illustrated in Fig. 8 is fairly repre- 
sentative of conditions in Hollywood. Some laboratories had sufficiently good 
developer agitation so that the spread between the curves was very much less than 
that shown here. Other sets of data showed the presence of more directional 
effect than is evident in Fig. 8. 

(5) The results of such tests depend strongly on the manner of reading test 
strips. Investigation showed that even a 2.4-mm diameter dot of fairly high 
density showed a considerable change in density if scanned across its diameter in 
the direction of film travel, providing it was developed in a machine with rela- 
tively poor agitation. Consequently, the dot densities reported here are the read- 
ings of the leading edges, and, therefore, the maximum densities of the dots when 
scanned with a viewing aperture, square in shape, and 0.62 mm on an edge. The 
Western Electric RA1100A densitometer with visual filter and a specially adapted 
aperture was used for these density readings. Tests made with larger dots of 
diameter 4.8 mm showed that a single dot of such size on a high gamma infinity 
film showed a variation in density across its diameter in the direction of film travel 
of from 1.12 to 1.37, while 1373 showed a variation of from 1.21 to 1.28. It is 
obvious that such lack of uniformity will not be encountered in scanning a single 
density step in a lib sensitometric tablet, for the reasons that this step has other 
steps adjacent to it which tend to erase the large variation found in a small area of 
high density immediately surrounded by clear film. However, in a variable- 
density sound track, it is likely that alternate areas of such size and density differ- 
ence do occur, so that wave-form distortion would be extremely great in the case 
of a high gamma infinity film coupled with poor developer agitation. Emulsion 
1373 will do much to alleviate this condition even under the poorest conditions of 
agitation. Incidental to the more general problem, this property of 1373 will 
greatly aid in the elimination of sprocket-hole modulation. 


The developed 1373 image has been found to have a lower color 
coefficient than other sound films, and the change in image color with 
increase in degree of development is less for 1373 than for other films. 
This is believed to be a result of the proximity to gamma infinity 
attained in ordinary development of this film. 

In the past it has been felt necessary to have a sound film that was 
capable of being developed to a high gamma so that the short ends 
and out-takes could be used for prints. Many people have believed, 
however, in so doing the best results for the purpose for which the 
film was designed, namely, sound recording, were not obtained. The 
results obtained with Type 1373, which is not capable of being de- 
veloped to a high gamma, bear out that contention. 

Type 1373 emulsion is very much finer grained than 1301 or 1357, 
and is correspondingly slower in emulsion speed. It has been found 


FIG. 9. Spectrogram of 1373 to tungsten light. 

by comparison tests, however, that 1373 requires less exposure in the 
sound recorder at conditions of normal use than any previous fine- 
grain sound recording emulsion. 

The color sensitivity of 1373 is shown in the wedge spectrogram in 
Fig. 9. This film, being blue-sensitive only, may be safely handled 
under a Wratten series or OA safelight. 

Emulsion 1373 is coated on a clear base, of the same type as used 
for emulsions 1357, 1301, and 1302. It is available in 16-mm width 
on acetate base, and carries the code number 5373. It is believed 
that this film will find widespread usage in the 16-mm variable- 
density recorders now being put into use by the motion picture in- 
dustry for rerecording 35-mm feature pictures for subsequent 16-mm 

Type 1373 film has the same over-all thickness as other variable- 
density sound recording films. Likewise, the emulsion thickness is 
about the same as 1357 and 1301. Nevertheless, it is suggested that 
focusing tests be made for maximum resolution when this emulsion is 



Vol 45, No. 4 

compared with others, since it has been observed that this usually 
achieves a slight gain in high-frequency response. 

Sound Quality Test Results. Light valve gamma measure- 
ments, as well as high-intensity sensitometric curves made in the 
manner previously described in reference to 1372, are shown for emul- 
sion 1373 in Fig. 10, along with a lib curve which was given the same 
development. The light valve gamma curve was obtained by chang- 
ing the bias on the light valve by measurable amounts. The differ- 
ence in mercury arc and tungsten exposure characteristics at high 
intensity is slight. The lib control gamma matches the light valve 
gamma for this emulsion. This has been confirmed by other tests at 
lower and higher values of gamma. Since the absolute values of ex- 



FIG. 10. Characteristic curves obtained on emulsion 1373 
with various types of exposure. All exposures received the same 

posure for the high-intensity curves are not known, the relative spac- 
ing of these curves on the log R axis is not significant. 

Emulsion 1373 shows improved high-frequency response when 
compared with 1301 or 1357. The signal-to-noise ratio is higher by 
approximately 6 db for 1373 than for 1301. These advantages are 
to be expected of a film capable of higher resolution. 

The intermodulation curves given in Fig. 11 were obtained by re- 
cording at a peak modulation of 2 db below valve clash, which is equal 
to approximately 80 per cent modulation. The 1000-cycle note was 
recorded at a level 12 db lower than the 60-cycle note. The numeri- 
cal value of the minimum distortion is a function of the type of ana- 
lyzer circuit used in .making the measurements. The low levels of 
intermodulation distortion; such as those shown in Fig. 11, are be- 
lieved to be as much the limiting values obtainable with the recording 

Oct., 1945 


and analyzing equipment itself, as they are indicative of the film be- 
havior. The latitude of print densities which can be used with a 
reasonably low value of distortion has been widened by the use of 

It is known that the required value of negative lib gamma for 
minimum intermodulation at fixed negative and print density levels 


<J 16 




White Light 

.40 50 .60 .70 .80 


FIG. 11. Intermodulation as a function of print density. East- 
man Fine-Grain Sound Recording Film, Type 1373. Standard 100- 
mil track, recorded in W E 100 A A recorder, 9.0-amp lamp, no filter. 

U V 



8.0 amp 



Tung. + 2 mm 
Corning 584 filter 

lib gamma 
Unbiased density 
Exposure (lamp current) 

8.6 amp 

7 If) Ifi n 

lib gamma 


varies considerably with laboratories. This may be traced, in part, 
it is now believed, to the large differences in developer agitation at 
the various laboratories. This is equivalent to stating that the lib 
control gamma does not bear the same proportionate relationship to 
the effective intermodulation test negative gamma at one laboratory 
as at another. This point will require direct proof, but is indicated 
by such data as presented in Fig. 8. 

The authors express appreciation for the many helpful suggestions 


and test data supplied by the personnel of studio sound departments 
too numerous to mention by name. Especial thanks are due Robert 
Hufford who prepared the diagrams used in this paper. 


1 DIMMICK, G. L.: "Improved Resolution in Sound Recording and Printing 
by the Use of Ultraviolet Light," /. Soc. Mot. Pict. Eng. t XXVII, 2 (Aug., 1936), 
p. 168. 

2 SANDVIK, O., AND SILBERSTEIN, G.: "The Dependence of the Resolving 
Power of a Photographic Material on the Wave Length of Light," /. Opt. Soc. 
Amer. and Rev. of Sci. Instr., 17 (Aug., 1928), p. 107. 

3 MEES, C. E. K.: "The Theory of the Photographic Process," The Mac- 
millan Co. (New York), 1942. 

4 O'DAY, D.: "Comparison of Variable-Area Sound Recording Films," J. Soc. 
Mot. PicL Eng. t 45, 1 (July, 1945), p. 1. 

6 FRAYNE, J. G., AND PAGLIARULO, V.: "The Effects of Ultraviolet Light on 
Variable- Density Recording and Printing," /. Soc. Mot. Pict. Eng., XXXIV, 6 
(June, 1940), p. 614. 

6 IVES, C. E., AND JENSEN, E. W.: "The Effect of Developer Agitation on 
Density Uniformity and Rate of Development," /. Soc. Mot. Pict. Eng., XL, 2 
(Feb., 1943), p. 107. 

7 LESHING, M., INGMAN, T., AND PIER, K. : "Reduction of Development 
Sprocket-Hole Modulation," /. Soc. Mot. Pict. Eng., XXXVI, 5 (May, 1941), 
p. 475. 

8 LESHING, M., AND INGMAN, T.: "Some Turbulation Characteristics of New 
Twentieth Century-Fox Developing Machine," /. Soc. Mot. Pict. Eng., 44, 2 
(Feb., 1945), p. 97. 


Summary. A description is given of the characteristics of 2 new du Pont fine- 
grain sound films. Type 232 is a positive used for white-light printing in conjunction 
with high gamma variable-density sound track negatives. Type 236 is a fine- grain 
recording negative film adaptable to both low and high gamma variable-density 
sound recording techniques. 

Standard practice for the majority of the variable-density record- 
ing system users had been, for several years prior to January 1, 1945, 
to print all production and release tracks with ultraviolet light. All 
controls and techniques had been standardized on this basis and mil- 
lions of feet of high gamma sound and music negatives had been ac- 
cumulated in libraries. 

The decision of several of the studios to discontinue ultraviolet 
printing as of the above date affected all phases of sound production : 
dailies, dubbing, and release. In addition, it posed the question of 
what to do with the extensive library material that required ultra- 
violet printing when used in combination with normally processed 
fine-grain positive films. 

A survey of the general situation indicated that there was no alter- 
native to reducing the contrast of the release negative to compensate 
for the increase in printing contrast with white light. However, all 
production negatives could be kept at the standard "high" gamma 
level and the library question could be solved if fine-grain white-light 
prints of suitable contrast could be made conveniently. 

Although fine-grain release positive processed in picture negative 
developers met contrast requirements, this approach was ruled out by 
practical laboratory considerations. 

Appreciating the urgency of the situation, the du Pont Company 
undertook to make available a fine-grain positive film which, when 

* Presented May 17, 1945, at the Technical Conference in Hollywood. 
** Photo Products Dept., E. I. du Pont de Nemours and Company, Inc., Holly- 




Vol 45, No. 4 

processed under standard release print conditions, would duplicate 
closely the effective contrast of ultraviolet printed fine-grain release 

Fortunately, much earlier tests had shown that the well-known du 
Pont Type 228 fine-grain master positive emulsion had approximately 
the desired effective contrast in positive developers, and could meet 
emulsion speed, signal-to-noise, and other criteria, very satisfactor- 
ily. This emulsion, with minor changes, was therefore made avail- 
able to the industry in December, 1944, as a fine-grain sound positive 
under the type number 232. 



D P 

FIG. 1. (1} Type 225 white-light print; 
gamma = 2.30. (2} Type 225 ultraviolet 
print; gamma = 1.55. (3) Type 232 white- 
light print; gamma = 1.50. 

Current practice in the studios effected by the change in printing 
technique is to develop all variable-density production negatives to 
the same contrast level as had been used for years with those intended 
for ultraviolet printing, and to make all daily, dubbing, and library 
prints with white light on Type 232. Laboratory processing of these 
prints is identical with that given fine-grain release prints. 

The close agreement in contrast and curve shape between ultra- 
violet printed released positive and white-light printed Type 232 is 
shown in the print-through-gamma curves of Fig. 1. These are not 
placed in their respective log E positions but serve to illustrate the 

Oct., 1945 



point. They show also the extent of the contrast decrease obtained 
with ultraviolet printing of fine-grain release positive Type 225. 

Rate of Development of Type 232. The time-gamma character- 
istics of Type 232, as processed by a representative Hollywood labo- 
ratory, are illustrated in Fig. 2. It will be noted that the rate of 
change of gamma with time is low, a factor which favorably in- 
fluences not only the accuracy of control but, more important, the ex- 
tent of 96-cycle modulation and associated development effects. 

Intermodulation tests, using the methods of Frayne and Scoville, 1 
have been employed to determine optimum control gamma values for 
Type 232 and, as will be shown later, for Type 236. The negative used 
in all cases, is a single track on which a 60-1000-cycle signal has been 



TYPE: 232 



00" J' 


FIG. 2. 

a jo' 

exposed in a Western Electric D 867 15 recorder with 80 per cent 
modulation of the light valve. 

Such tests, using a high gamma negative with white-light printing, 
have indicated a control gamma of approximately 1.6 as most suit- 
able for Type 232. This value is readily obtainable in release positive 

Intermodulation curves, such as shown in Fig. 3, indicate the 
distortion and print density latitude to be at least the equivalent of 
the best obtainable from ultraviolet prints from the same class of 
negatives made on fine-grain release positive. 

No accurate data are available at the present time on the signal-to- 
noise ratio of Type 232 white-light prints, but the experience of several 
months in production has shown them to be definitely quieter than 
ultraviolet prints on fine-grain release positive. Some of the over- 
all improvement may be a result of reduced image graininess, but a 



Vol 45, No. 4 

substantial proportion originates from their carrying appreciably less 
printed-through dirt and abrasion noise from the negative. Regard- 
ing the latter, all evidence points to the primary factor being the 
change in printing light color rather than to optical or mechanical 
alterations in the printing machines. The reduced noise level in 
Type 232 dubbing prints, of course, carries through to the final re- 
lease product as a definite improvement. 

Comparative frequency response measurements show a minor 
advantage of less than 1 db at 7000 cycles in favor of ultraviolet 
printed release positive. 

FIG. 3. 


Negative: Type 226, gamma 0.85, density 0.55. Print 
Type 232, gamma 1.63. 

In considering the release print phase of the shift from ultra- 
violet to white-light printing, it was obvious that the contrast of the 
release negative would have to be lowered to compensate for the 
increase in contrast encountered with white-light printing. Inter- 
modulation tests indicated approximately a 30 per cent decrease in 
negative control gamma which, for a given negative density, re- 
quired more than a one-ampere increase in recorder lamp current. 

Du Pont Type 226 fine-grain recording film, designed originally for 
high gamma recording purposes, proved too low in emulsion speed 
to meet the new requirements when used in standard variable-density 
recording equipment. Therefore a second and more sensitive fine- 
grain recording film, Type 236, was made available prior to the time 
of the studio change in printing methods. 

Careful consideration was given in its design to both low and high 

Oct., 1945 



gamma recording requirements. The emulsion speed, for example, 
was made just sufficient for low gamma recording purposes so that 
grain noise might be held to minimum values. Contrast factors were 
adjusted for convenient operation in developing solutions already 
available in Hollywood laboratories. 

Development Characteristics of Type 236. Fig. 4 shows a lib 
and 2 time-gamma curves for Type 236, as processed in a Hollywood 
laboratory which employs an MQ developer similar in formula to the 
du Pont ND4. 


FIG. 4. 

The lower of the 2 time-gamma curves is representative of the 
laboratory's standard operation and shows a low rate of change of 
gamma with time; this is insurance of minimum 96-cycle modulation 
and other undesirable development effects. 

The upper curve illustrates the effect on Type 236 of an 0.12 in- 
crease in the pH of the formula a simple and effective means of 
adjusting the activity of this class of MQ developers. 

Intermodulation tests on the release combination of low gamma 
Type 236 negative, white-light printed to fine-grain positive, have 
indicated that a negative control gamma of approximately 0.65 is 
required with the standard Western Electric D86715 recorder. This 
seemingly high gamma value is explained by the fact that the actual 



Vol 45, No. 4 

light- valve gamma is appreciably lower than the control gamma. 
Light- Valve and lib Characteristics of Type 236. A comparison 
between light- valve and lib sensito metric exposures on Type 
236, given identical processing, is shown in Fig. 5. The gamma of 
former is 0.47 and of the latter, 0.63. The relationship between light- 
valve and lib gammas is so constant that the latter can be used 
satisfactorily for all control purposes. However, recorder optical 
arrangements can affect the gamma differential; this should there- 
fore be checked either by light- valve or intermodulation tests for each 
type of recorder. 

L V GAMMA -O.4-7 


FIG. 5. 

Intermodulation from White-Light Print on Type 225. A repre- 
sentative intermodulation curve for the release combination of Type 
236 white-light printed to Type 225 release positive is shown in Fig. 6. 
Satisfactory distortion values and good printing latitude are in- 

Critical listening tests, which effectively integrate factors such as 
frequency response and signal-to-noise ratio, have established that 
no loss in sound quality has accompanied the change to white-light 
release printing with Type 236 employed as the release negative. 

Type 236 has not as yet been applied extensively to production 
recording at high gamma levels wherein it would replace the standard 
Type 226. However, data indicate that it is fully the equal of the 
latter in respect to distortion and print latitude and that, because of a 
lower rate of development, it will show appreciably less 96-cycle 

Oct., 1945 



modulation if processed under conditions of poor developer agitation. 
Intermodulation from White-Light Print on Type 232. An 

intermodulation curve for Type 236 at high gamma levels, white-light 
printed to Type 232, is shown in Fig. 7. Distortion and print latitude 
characteristics are fully as good as in the other combinations covered 

It can be admitted now that the prospect of returning to white- 
light printing of variable-density negatives was not viewed opti- 
mistically when first considered. This was quite natural, since 
intermodulation records from the period prior to the change-over to 

.40 .SO .60 JO 


FIG. 6. Intermodulation from white-light print on Type 225 
Negative: Type 236, gamma 0.64, density 0.52. Print: Type 225, 
gamma 2.30. 

ultraviolet printing showed distortion values several-fold as great as 
with the more recent ultraviolet printing of fine-grain films. These 
data, however, were based on the relatively coarse-grained recording 
and printing films of that period and have been proved unnecessarily 
alarming by the original tests and months of production experience 
with the white-light printed combinations of du Pont Type 232 and 
236 fine-grain sound films. 

Intermodulation of Release Prints. Intermodulation curve 3 of 
Fig. 8 is representative of current release practice wherein a Type 
236 low gamma negative is white-light printed to Type 225 release posi- 
tive; it is essentially identical in distortion and latitude with the 
curves shown earlier for the other white-light printed combinations 
of the new films. Comparison of curve 3 with curve 2, which is 



Vol 45, No. 4 

representative of high gamma negatives ultraviolet printed to the 
same release positive, shows an advantage to the former in both dis- 

2 - 


,4O .SO .60 >'O 


FIG. 7. Intel-modulation from white-light print on Type 232 
Negative: Type 236, gamma 0.76, density 0.63. Print: Type 232, 
gamma 1.55. 


FIG. 8. Intermodulation of release prints. Curve (1): Old- 
type recording and positive films printed with white light. No 
data. (From Frayne and Scoville. 1 ) Curve (2): High gamma 
negative printed with ultraviolet. Negative: Type 226, gamma 
0.64, density 0.55; Print: Type 225, gamma 2.30. Curve (3): 
Low gamma negative printed with white light. Negative: Type 
236, gamma 0.64, density 0.52; Print: Type 225, gamma 2.30. 

tortion and print latitude, particularly in the useful low print density 


Curve 1, from the paper of Frayne and Scoville, 1 may be considered 
as representative of best practice with the old-type recording and 
positive films, white-light printed, and is included as a matter of 
interest to show the extent of the improvement obtained with fine- 
grain films. 

It may be said, in conclusion, that the change from ultraviolet to 
white-light printing has been effected without loss in sound quality 
or in facility of operation with the assistance of du Pont fine-grain 
sound films, Types 232 and 236. 


1 FRAYNE, J. G., AND SCOVILLE, R. R.: "Analysis and Measurement of Dis- 
tortion in Variable- Density Recording," /. Soc. Mot. Pict. Eng., XXXII, 9 
(June, 1939), p. 648. 


Summary. A brief description is given of the establishment and activities of the 
U. S. Naval Photographic Services Depot in Hollywood. The Depot serves as the 
production center for all of the Navy's motion picture training films made on the 
West Coast. 

The U. S. Naval Photographic Services Depot, at 1357 North Vine 
Street in Hollywood, California, is the West Coast station of the 
Photographic Division of the Navy's Bureau of Aeronautics. It was 
established on June 23, 1943, by the late Secretary of the Navy Frank 
Knox. Prior to tjiat the Navy's then limited Hollywood activities 
had been conducted from offices in the Disney Studios. 

The depot is charged with handling all of the Navy's picture busi- 
ness on the West Coast, with emphasis on the Hollywood area. Its 
largest responsibility is the production of training films for the 
bureaus and commands of the naval establishment. These pictures 
are produced by the major studios, by the animation and model 
people, by the small independents who made business films in peace- 
time, and by Navy and Marine Corps camera units. 

The production load has been averaging around 100 films. In 1944 
the station completed 96 pictures, many of them as complicated as 
theatrical features. The activity's officer-in-charge refers to his sta- 
tion as "the feature department." Thirty-one of these were pro- 
duced from script to screen by Navy and Marine units in 31 weeks. 
The subjects were the facilities, doctrine, and personnel of amphibious 

The major studios produce for the Navy at the actual cost of doing 
the picture. The animation and model studios and the business film 
producers receive from 7 to 10 per cent profit, plus overhead. The 
Navy and Marine crews produce for just what it costs to keep them 

* Presented May 17, 1945, at the Technical Conference in Hollywood. 
** Lieut. Commander, USNR, Officer-in-Charge, U. S. Naval Photographic 
Services Depot, Hollywood. 


going. This military personnel production is averaging $1.55 per ft, 
a figure which has held for government-employee picture making for 
many years. The 4 camera crews functioning at the depot usually 
are reserved for highly classified, high priority projects, involving a 
volume of naval personnel and equipment. Usually these pictures, 
which must be completed and released in a few weeks, are concerned 
with new weapons and doctrines of modern warfare. 

The majors get the "features," the 6 or 8 reelers requiring extensive 
production, usually the indoctrination or "attitude establishing" type 
of film. The little independents take on the simpler jobs, and the 
animation-model people get the animation and the models. If a film 
being made by a studio contains sequences which call for authentic 
background material which can be shot only on location at some 
naval establishment, these sequences will be assigned to Navy camera 
crews. If desirable, the studio director will take over the Navy 
crew and cut in the sequences after completion of the studio footage. 

By virtue of a unique contract with Paramount Pictures, the sta- 
tion's Navy camera crews have access to the stages and facilities of 
that company. And the local labor groups have cooperated in work- 
ing side by side with men in uniform. 

Production jobs are assigned to the station at various stages. 
Some bring no more information than the title and the source of 
technical advice. Others come as production outlines, and others 
arrive as completed scripts. Nearly all writing is done by personnel 
of this depot. No project is taken to a major studio except in shoot- 
ing script form. That is the way they like it. 

The station maintains a procurement section which solicits pro- 
duction proposals from producers and gets the budgets in proper 
form prior to sending to Washington for processing and final approval. 
This procedure has eliminated much production delay. 

Production is the depot's number one activity, but there is some 
distribution of color prints and prints which must go direct to the 
fleet or West Coast shore establishments. Certain special services 
also come into the picture. These may include repair of projection 
equipment or cameras for a ship or shore facility, experimental 
photography, shooting a few hundred feet on some rocket test for 
Cal. Tech., the production of some localized short for a particular 
command, and many other odd jobs. Recently the station designed 
a wing camera which may well revolutionize the whole business of 
aerial photography. 

296 F. M. HEARON 

The premises of the station include individual offices, a shooting 
stage, stowage and shipping quarters, 2 projection rooms, a still 
laboratory, Moviola room, art department, special effects department, 
and the inevitable coffee mess. 

The personnel at the depot are largely from the motion picture 
industry. The current complement (increases have been requested) 
is 27 officers and 42 enlisted men, but there usually are around 100 
persons aboard. The extras are photographic personnel and tech- 
nical advisors on temporary duty. These technical advisors for the 
various types of films are among the Navy's most competent officers. 
They have studied their subjects, seen them in action, and now it is 
their job to get them on film so others may learn easily what they 
have learned the hard way. 


Summary. Described in this paper is an optical device of simple arrangement 
which serves to permit accurate placement of pickup stylus for a series of cues in 
photography to music playback from disk. 

Various applications of disk records and transcriptions are facili- 
tated by the use of a cueing or spotting device which will enable ac- 
curate placement of the pickup stylus on the disk record. In fact, 
if extensive versatility is to be realized in the application of standard 
pitch disks to photography by playback, a highly accurate device for 
cueing is essential. Some musical productions involve continuous 
sequences of photography, in which live sound pickup with dialogue is 
alternated with playback of music from disk recordings. For ex- 
ample, such a sequence may start with introductory dialogue be- 
tween vocalists, to be followed by playback music which must start 
precisely on a word cue. Then, with the camera continuing to cover 
the same scene, the vocalists may go directly from action with play- 
back into live dialogue, followed by a cue for another vocal selection 
or dancing to playback. Still further alternations between live sound 
pickup and action to playback may follow in the same scene, requir- 
ing that the playback operator be able to start each playback section 
accurately on cue. 

A number of cueing aids, ranging from simple marked strips of 
cardboard to rather intricate mechanical-electrical devices, have been 
made and used in radio and motion picture studios. Some of these 
devices have undoubtedly rendered satisfactory service in the ap- 
plication for which they were specifically designed. However, local 
requirements created the need for a device of simple construction 
which would permit operation with split-second cueing. It was de- 
sirable that this device should not involve attachments of appreci- 

* Presented May 16, 1945, at the Technical Conference in Hollywood. 
** Major, Signal Corps, Signal Corps Photographic Center, Long Island City, 




Vol 45, No. 4 

able weight to the pickup arm, nor should apparatus be installed over 
the turntable which would impede the handling of disks. Of prime 
importance was the requirement that the device enable pre-setting 
accurately as many cues as desired for a scene of photography, with 
reliable performance in order to preclude the necessity of retakes 
caused by errors in playback cueing. The construction was to be 
such that maintenance would not present a problem. 

FIG. 1. 

Schematic plan view of arrangement of play- 
back cueing device. 

The arrangement arrived at and now in use is indicated in Figs. 
1 and 2. The optical system is essentially that of certain variable- 
area recorders up to the point of the recording mask image on the slit. 
A one-ampere exciter lamp illuminates a hairline object, the latter 
being mounted on a cover glass in front of a condenser in the lamp 
house. The condenser images the lamp filament in a small mirror 
mounted on the pickup arm directly above its pivotal point. Mounted 
on a post near the pickup arm and mirror is a simple-element objec- 
tive lens. This objective forms an image of the hairline on a slide 
and rail assembly at the opposite side of the turntable cabinet, with 
the beam reflected by the mirror in the image space. A single-ele- 


ment objective with a relative aperture of approximately f/8 pro- 
vides sufficient image brightness, if the rail is shielded from spurious 

As represented by the traverse of the hairline image, the lateral 
movement of the pickup stylus is magnified by a factor of two 
through the mirror reflection. The ratio of the slide-mirror distance 
to the stylus-mirror distance determines the additional magnification 
obtained. The total magnification in the units constructed is ap- 
proximately five. For this factor and with a disk cut to a pitch of 100 
lines per in., the hairline image will move 0.050 in., or 1.27 mm per 
revolution of the disk, an easily discernible increment. If the disk is 

FIG. 2. Sketch showing light beam focused on slide index. 

made on a recorder which has cutter feed error, the resulting eccen- 
tricity of the cut will produce a nonuniform motion of the hairline 
image. However, since each cue will be taken for a given angular 
orientation of the turntable, the eccentricity will cause no difficulty 
in using the device. This follows from the fact that for any given 
orientation of the disk, the increments of stylus traverse per revolu- 
tion are essentially uniform. 

In operation, the slides are set for each cue by running in while 
listening, and stopping the disk with the hand while a slide index line 
is aligned with the hairline image. The playback units are equipped 
with headphones so that the operator may pre-set the cueing device 
without disturbing the stage. Once the cues are pre-set, the opera- 
tor may return to any one of them quickly and as many times as de- 
sired. This facilitates rehearsals in particular. The cues may be 

300 G. C. MlSENER Vol 45, No. 4 

applied in various ways. For example, they may be used to indi- 
cate the points at which the playback signal is to be switched or 
faded, on or off, while the disk is running. In other cases, the cues 
may be used to indicate when the pickup is to be dropped to or lifted 
from the running disk. In still other instances, the disk may be held 
by hand, with the turntable rotating, and the pickup placed by aid of 
the cueing device. The disk is then released on an action cue from 
the stage. 

To facilitate placement and release of the pickup arm, an auxiliary 
device is mounted under the pickup arm near the turntable, as shown 
in Fig. 2. This consists of a horizontal bar suitably mounted, which 

FIG. 3. Photograph of unit in use at Signal Corps Photographic Center. 

is raised and lowered by means of a hand-operated cam. Lateral 
adjustment of this bar support is provided by a thumbscrew. With 
this arrangement, the pickup arm may be placed on the bar so that 
the hairline image is approximately aligned with a slide index line. 
Then the thumbscrew is used as a vernier adjustment to secure exact 

The device described is obviously simple in construction (see Fig. 3) 
and requires little maintenance, yet offers the advantage of allowing 
pre-setting for as many cues as may be necessary in a scene of pho- 
tography to playback. There is nothing to obstruct changing or 


reversing disks, and the only apparatus attached to the pickup arm 
is a small mirror of inconsequential weight. The pre-set cues may 
be applied either with the disk in motion or held in position. 

This arrangement suggests other variations which might be help- 
ful in meeting special requirements. For example, the indexed 
slides might be provided with a slit, behind which a photocell or 
cells could be mounted for automatic activation of audio-control 
circuits or signal devices on the stage. In cases where space for pro- 
jection of the hairline image is limited, adequate magnification of the 
stylus traverse may be obtained by placing a second mirror on a fixed 
mount near the mirror on the pickup arm. This will provide an ad- 
ditional factor of two in the magnification of the angular displace- 



Summary. A description is given of the method devised by the authors and 
used at the Signal Corps Photographic Center to eliminate human error in critical 
focusing by automatic means. The fully-automatic dolly was designed primarily 
to eliminate these difficulties when the camera is in horizontal movement. 

In the course of producing training films at the Signal Corps Photo- 
graphic Center, Long Island City, New York, it was soon realized by 
the cinematographers attached to the Camera Branch that, in many 
instances, the type of photography required for training films, orien- 
tation films, morale films, etc., was somewhat different than the pho- 
tography required for the production of entertainment films as 
normally produced in the major Hollywood studios. At the outset 
the production cameramen and special effects cameramen soon found 
that a great percentage of the Signal Corps productions required 
numerous shots of maps, diagrams, mock-up models, inserts, minia- 
tures, etc. In so many shots of this type it was necessary to open the 
scene with a full shot of a map, diagram, or model, and then move the 
camera in to a specific point in order to call attention to it or em- 
phasize it. The reverse of this procedure was also often the case, 
where it was necessary to open the shot at a specific point and then 
move the camera back in order to encompass the entire object. 

Many difficulties were encountered in an effort to photograph such 
shots with the required degree of accuracy. First of all, the Camera 
Branch was hampered by having only a few highly trained and com- 
petent camera operators capable of operating the camera in these 
extremely difficult shots and also only a few assistant cameramen 
who had had enough experience in changing focus on lenses accu- 
rately. The procedure of accurately changing focus on a photograph- 

* Presented May 16, 1945, at the Technical Conference in Hollywood. 
""^Captains, Signal Corps, Camera Branch, Signal Corps Photographic Center, 
Long Island City 1. N. Y. 



ing lens while the camera is in motion is extremely difficult and re- 
quires a great deal of instruction and constant practice. It was 
found that inexperienced camera operators were never sure if the 
object being photographed was in sharp focus throughout the entire 
length of the scene, and it was always necessary to rephotograph the 
scene several times with the hope that one of the "takes" would be in 
sharp focus. 

FIG. 1. Automatic dolly. 

Difficulties were also experienced with camera dolly "weave" and 
vibrations, resulting from the human element introduced in start- 
ing and stopping the dolly and variations in the dolly tracks. For the 
same reason, acceleration surges were seldoma bsent. Also, accu- 
rate synchronization of in-and-out movements with up-and-down 
movements of the camera was seldom realized. 

It soon became apparent that it would be advantageous to con- 
struct some sort of device which would eliminate the human element 
not only in moving the dolly but also automatically changing the focus 

304 J. T. STROHM AND W. G. HECKLER Vol 45, No. 4 

of the photographing lens during the periods when the camera was in 
horizontal movement. Research work was begun to accomplish 
this end and resulted in the development of the present all-electric 
and fully automatic camera dolly which is used at the present time by 
the Camera Branch at the Signal Corps Photographic Center. 

This fully automatic electric dolly is an adaptation of a standard 
Raby camera dolly. (See Fig. 1) The 2 standard rubber wheels 
are retained on the left side, while the standard rubber wheels on the 
right side have been replaced with 2 bronze wheels which have had a 
F-shape groove cut into their riding surfaces. Round l /z-in.. tubings 
which are countersunk into a wooden base act as a straight-line guide 
for these wheels. Troublesome weave and vibrations are completely 
eliminated by this new combination of dolly guide wheels and track 
tubing. The track joints themselves are carefully butted together, 
eliminating the usual track irregularities. 

The power unit which motivates the dolly consists of a l /*-hp, 
110-v d-c motor which has a top speed of 1725 rpm. The shaft of 
the motor is connected to the speed reducer box by means of a rubber 
coupling that takes up all motor vibrations as well as start and stop 
jars. The speed of the d-c motor is reduced 50 times by the reducer 
box, and a sprocket gear pulley from it engages the sprocket chain 
which in turn rotates the dolly axle and bronze power wheel. A 
rheostat governs the speed control, while standard reversing switches 
determine direction. 

Essentially, the same type of power unit has also been installed 
on the dolly and is applied to the dolly tilt arm so that it may be 
raised or lowered with ease even when the dolly is in motion. To the 
rear of the unit a seat for the dolly operator is provided, together with 
a control panel which contains switches and controls that govern the 
speed and movement of both the dolly and the tilt arm. The control 
panel also contains dials which show any given position of the dolly 
and the tilt arm. Once the dolly operator knows the conditions of 
the shot he can duplicate these conditions any number of times 
without fear of error, for any error that he might make would be 
plainly indicated on the control panel after the shot was completed. 
Also, additional switches make it possible for the operator to con- 
trol the entire series of movements by the throwing of a master 

The most important mechanism installed on the dolly, however, 
is an automatic follow-focus device. As stated above, this device 



was developed and installed because of the great need for accurate and 
positive focus, particularly on close follow shots where the narrow 
depth of field characteristic of photographic lenses as they closely 
approach a given target demands extremely accurate focusing. This 
need was of particular importance in special effects work where fol- 
low shots are concerned mainly with extremely accurate framing and 
the extreme proximity of the lens to the object or target. Such ex- 
amples can be cited as the need to move from a close-up of an indi- 
vidual to his mouth or eyes, or in some cases, to one eye. Another 

FIG. 2. Dolly cam and gear assembly. 

common case could be cited such as moving up to or away from small 
sections of maps or titles. 

Although this device could be used in many instances in standard 
set procedure, no intent was made to displace current production 
methods and it was conceived only for those highly difficult follow 
shots which are almost impossible to accomplish when the camera- 
man must depend upon the judgment of the operator or assistant to 
focus the lens by hand. The automatic focusing of the photograph- 
ing lens is accomplished in the following manner. 

The focus unit receives its activation from the right front dolly 
wheel. (See Fig. 2) It transfers this energy to a cam, which has a 
contour pitch, complementary to the curvilinear action of a 2-in. 


J. T. STROHM AND W. G. HECKLER Vol 45, No. 4 

lens or of the particular focal length lens desired. This action is 
applied to a small gear on the end of the shaft of a Selsyn generator 
motor. The rotation of this motor is transmitted to and received 
by a Selsyn receiving motor. A small gear, same size as on the gen- 
erator, is mounted on the end of the receiving motor shaft. This 
activates a pinion gear, which turns the actual lens gear itself. (See 
Fig. 3) 

There is a distinct advantage in using electrically connected Selsyn 
or interlock-type motors to transmit the movement of the dolly to the 
photographing lens. As can easily be seen, this does not restrict 
the movement of the camera in any way. The only connection be- 
tween the receiver motor and 
lens assembly, and the motor 
which is activated by the cam 
and gear assembly, is a flexible 
cable containing only the 
motor wires. Hence, it can 
be seen that the camera may 
be tilted up or down or panned 
to right or left without any 
hindrance whatsoever. 

A standard Mitchell camera 
is used on the unit, which is 
equipped with a 50-mm 
Bausch & Lomb Baltar, f/2.3 
lens in a standard Mitchell 
lens mount. To the lens 
mount a ring gear was 
mounted which is meshed 
with the control gear of the 

receiver motor assembly. The 50-mm lens can be focused auto- 
matically from 50 ft to 18 in. Within these limits, no matter where 
the dolly is moved or at what speed it is moved, the lens is always 
automatically held in sharp focus. 

The benefits derived from this unit are numerous. One advantage 
of its use has been a great saving in both time and labor. Before 
the unit was in operation it was necessary to use as many as 6 men to 
complete a difficult follow shot. In some instances scenes of this 
nature required a camera operator, an assistant cameraman to change 
focus, one or 2 men 1^o push the dolly, a fifth man to call out footage 

FIG. 3. Dolly lens gear assembly. 


marks, usually marked on the floor, and possibly a sixth man to carry 
the camera motor cable back and forth as the dolly was moved. As 
mentioned before, the common practice was to photograph the scene 
many times hoping that at least in one of the "takes" all of the 
technicians connected with the scene had coordinated and synchron- 
ized their operations correctly. This, of course, required a great 
amount of time, an abnormal waste of film, and usually a crew of 
from 4 to 6 men. With the use of the automatic electric dolly most 

FIG. 4. Title and insert stand. 

of these disadvantages were eliminated. No matter how difficult 
the scene the unit requires the use of only 2 men, the camera operator 
and the dolly operator. The only function of the camera operator 
is to start and stop the camera and to operate the pan and tilt head 
if this should be necessary. The dolly operator controls the move- 
ment of the dolly, and all other necessary operations are performed 

Since the unit has been in use it has been found that there are few 
occasions where it is necessary to make more than one take of the 
scene. The saving in film because of this advantage can be recog- 
nized at once. 

308 J. T. STROHM AND W. G. HECKLER Vol 45, No. 4 

In conjunction with the development of the automatic follow-focus 
electric camera dolly, a similar device was developed to accomplish 
the same results on a permanently installed Insert and Title Stand. 
(See Fig. 4) In many cases it was found to be more convenient to 
mount certain maps, titles, and other special objects on a title board 
which was placed in an upright position and attached to a lathe bed. 
The camera was mounted on a movable pedestal which in turn was 
mounted on a smooth raceway. This raceway was substituted for 
the original lathe rack and constructed in such a way as to permit the 
camera to be moved back and forth on it. When the unit was put 
into use, approximately the same problems presented themselves as 
before. It was even more difficult to change focus accurately when 
moving the camera, for the shots made with this unit usually required 
ajhigher degree of accuracy both in focusing and framing. It was 
necessary to design the equipment in such a manner so as to allow the 
3-in. camera lens to approach the title board or target as close as 12 

Because of the complex nature of certain shots, it was also decided 
that there would be a distinct advantage in being able to move the 
title board automatically in either a horizontal or vertical direction. 
To accomplish these features, the unit was reconstructed in the 
following manner. 

The title board part of the installation is made to move in a hori- 
zontal and vertical direction by means of 2 Bodine speed reducer-type 
animation motors. One of these motors powers the horizontal 
movement, the other the vertical. The single frame feature permits 
small precise moves for straight or animation work. The reversing 
switches provide directional control. The speed adjustments allow 
speed control for board movements. Both motors are geared down 
by 12 to 1 reducer boxes. This smooths out the movements and gives 
proper basic speed. The follow focus is effected by mounting a con- 
tour strip, complementary to the linear movement of a 3-in. rack type 
of lens along the side of the lathe bed. A small ball-bearing roller 
makes contact with this contour. A shaft connects the roller bearing 
to the shaft of a Diehl-type Selsyn generator motor via reduction 
gears. The action of the ball bearing as it follows the contour strip 
activates the generator which electrically transmits identical turns 
to the receiving Selsyn motor's shaft. On the end of the shaft of 
this motor, a small gear engages and activates the rack and pinion 
gear directly attached to the rack lens mount itself. 


In the actual practice of cinematography at the Signal Corps Photo- 
graphic Center, both of these devices have been used with a great 
deal of satisfaction by the cinematographers who are charged with the 
responsibility of making these difficult shots. During peak periods 
of production they have enabled the Camera Branch to complete 
many different scenes of this type where formerly it was possible to 
complete only a limited number. 

It is felt that this dolly with the automatic focus device could be 
very successfully utilized in television camera operation because of 
its remote control and pre-set switch features. Another suggested 
use for this unit would be in connection with rear projection or proc- 
ess photography where it might be advantageous to dolly the pro- 
jector in or out during a scene. 

The writers wish to express their appreciation to the Pictorial En- 
gineering and Research Laboratory Division and the Central Ma- 
chine Shop Branch of the Signal Corps Photographic Center for their 
cooperation and valuable assistance in the design and construction of 
these devices. 



The editors present for convenient reference a list of articles dealing with subjects 
cognate to motion picture engineering published in a number of selected journals. 
Photostatic or microfilm copies of articles in magazines that are available may be 
obtained from The Library of Congress, Washington, D. C., or from the New York 
Public Library, New York, N. Y., at prevailing rates. 

American Cinematographer 

26 (Aug. 1945), No. 8 

The Academy War Film Library (p. 261) E. GOODMAN 

A Crumbled Movie Empire (p. 262) I. BROWNING 

Films in India (p. 265) F. BERKO 

Fades, Lap-Dissolves, and Other Tricks (p. 266) R. PALMER 
The Production of Scientific Films for Biological 

and Medical Purposes (p. 295) J. Y. BOGUE 

Shooting Production Under Fire (p. 296) H. A. LIGHTMAN 

Cradle of American Cinema (p. 298) I. BROWNING 
Cine-Chronized Sound on Wire for Amateurs 

(p. 300) L. CONWAY 


25 (Aug., 1945), No. 8 
Resonant Loudspeaker Enclosure Design (p. 35) F. W. SMITH 

General Electric Review 

48 (Sept., 1945), No. 9 
Image Contrast in Television (p. 13) C. H. BACHMAN 

International Photographer 

17 (Aug., 1945), No. 7 
Movie Sound to Order (p. 13) 

Alton Improved Dolly Track (p. 14) J. ALTON 

The Walker Comparison Meter (p. 18) J. WALKER 

G. E. Flask Tube (p. 26) 

17 (Sept., 1945), No. 8 

Photographing Air Routes and Installations (p. 9) F. CROSBY 

Making 16-Mm Reduction Prints (p. 12) S. A. COHEN 

Visual Music (p. 19) J. ALTON 

International Projectionist 

20 (Aug., 1945), No. 8 


Oct., 1945 



Projection Television (p. 7) 

Some Common and Uncommon Remedies for 
Hum in Sound Systems (p. 12) 

Projectionists' Course on Basic Radio and Tele- 
vision Pt. 14, Capacitance (p. 18) 

20 (Sept., 1945), No. 9 
A Wartime 16-Mm Projector: Its Operation and 

Maintenance (p. 7) 

Projection at Radio City Music Hall (p. 12) 
Screens : From a Sound Engineer's Point of View 

(p. 14) 

Projectionists' Course on Basic Radio and Tele- 
vision Pt. 15, Capacitance (Contd.) (p. 19) 
The Orthpscope Lens (p. 26) 

Technique Cinematographique, La 

15 (May 15, 1945), No. 2 

Alphonse Seyewetz (1869-1940) Obituary Notice 
Physiological Problem of Vision and the Motion 
Pictures (p. 27) 

15 (June 1, 1945), No. 3 
Masks and Diffusers in Taking (p. 45) 
A Registration System for 5-Mm Push-Pull 
(P- 47) 

15 (June 15, 1945), No. 4 

Suppression of Calcareous Deposits in the De- 
velopers and on the Developing Machines 
(p. 61) 

Modern Light Sources from the Point of View of 
Photography and Motion Pictures (p. 67) 












Designer and engineer experienced in optics, lighting, and microphotog- 
raphy, capable of designing microfilm reading equipment and products 
related to microfilm industry. Reply to Microstat Corporation, 18 
West 48th St., New York 19, N.Y. 

Design engineer, experienced in mechanics and optics of motion picture 
cameras, projectors, and film scanning. Give details. Reply to Mr. 
John H. Martin, Columbia Broadcasting System, Inc., 485 Madison 
Ave., New York 22, N.Y. 



Sound recording engineer, 16- or 35-mm equipment, studio or location 
work, single or double system. Free to travel. For details write J. J. K., 
354 Ninth Ave., New York 1, N.Y. 

Honorably discharged veteran with 15 years' experience in all phases of 
motion picture production, including film editing, directing, producing. For 
details write F. A., 30-71 34th St., Long Island City 3, N.Y. Telephone 
AStoria 8-0714. 

Projectionist-newsreel editor with 15 years' experience just released 
from service. Willing to locate anywhere. Write P. O. Box 152, Hamp- 
den Station, Baltimore 11, Maryland. 

Director of visual training aids, now concluding government work, de- 
sires position with educational, industrial or commercial organization, 
supervising or assisting in the production or distribution of visual training 
or allied work. Write H. C. B., 348 Maryland Ave., Dayton 4, Ohio. 

Chief Engineer of motion picture camera manufacturer now available. 
Special training in optics, electricity, electronics, mechanics. Experienced 
in all phases of manufacture of cameras, projectors, and accessories. 
Prefer West Coast, but not essential. Write Robert Winkler, 119 West 
78th St., New York, N.Y. 


Vol45 NOVEMBER, 1945 No. 5 



Machine Processing of 16-Mm Ansco Color Film 

J. L. FORREST 313 

Practical Utilization of Monopack Film 

C. G. CLARKE 327 

A Multisection Rerecording Equalizer W. L. THAYER 333 

An Improved Loudspeaker System for Theaters 


Reverberation Chambers for Rerecording 


Continuous Flash Lighting An Improved High-In- 
tensity Light Source for High-Speed Motion Picture 
Photography H. M. LESTER 358 

Automatic Recording of Photographic Densities 


Variable- Area Release from Variable- Density Original 
Sound Tracks J. P. LIVADARY AND S. J. TWINING 380 

Society Announcements 389 

Program of 58th Semi-Annual Technical Conference 391 

Copyrighted, 1945, by the Society of Motion Picture Engineers, Inc. Permission to republish 
material from the JOURNAL must be obtained in writing from the General Office of the Society. 
The Society is not responsible for statements of authors or contributors. 

Indexes to the semi-annual volumes of the JOURNAL are published in the June and December 
issues. The contents are also indexed in the Industrial Arts Index available in public libraries. 





Board of Editors 





Officers of the Society 
** President: DONALD E. HYNDMAN, 

350 Madison Ave., New York 17. 
** Past-President: HERBERT GRIFFIN, 

133 E. Santa Anita Ave., Burbank, Calif. 
** Executive Vice-President: LOREN L. RYDER, 

5451 Marathon St., Hollywood 38. 
* Engineer ini Vice-P resident: JOHN A. MAURER, 

37-01 31st St., Long Island City 1, N. Y. 
** Editorial Vice-President: ARTHUR C. DOWNES, 

Box 6087, Cleveland 1, Ohio. 
* Financial Vice-President: ARTHUR S. DICKINSON 

28 West 44th St., New York 18. 
** Convention Vice-President: WILLIAM C. KUNZMANN, 

Box 6087, Cleveland 1, Ohio. 
4 'Secretary: E. ALLAN WILLIFORD, 
230 Park Ave., New York 17. 
*Treasurer: M. R. BOYER, 
350 Fifth Ave., New York 1. 


*FRANK E. CARLSON, Nela Park, Cleveland 12, Ohio. 
**JOHN I. CRABTREE, Kodak Park, Rochester 4, N. Y. 
**CHARLES R. DAILY, 5451 Marathon St., Hollywood 38. 

*EDWARD M. HONAN, 6601 Romaine St., Hollywood 38. 
*JCLYDB R. KEITH, 233 Broadway, New York 7. 

*G. T. LORANCE, 92 Gold St., New York 7. 
**PETER MOLE, 941 N. Sycamore Ave., Hollywood. 
*fHoLLis W. MOYSE, 6656 Santa Monica Blvd., Hollywood. 
**WILLIAM A. MUELLER, 4000 W. Olive Ave., Burbank, Calif. 

*EARL I. SPONABLE, 460 West 54th St., New York 19. 
**REEVE O. STROCK, 111 Eighth Ave., New York 11. 

*WALLACE V. WOLFE, 1016 N. Sycamore St., Hollywood. 

*Term expires December 31, 1945. fChairman, Pacific Coast Section. 
**Term expires December 31, 1946. {Chairman, Atlantic Coast Section. 

Subscription to nonmembers, $8.00 per annum; to members, $5.00 per annum, included in 
their annual membership dues; single copies, $1.00. A discount on subscription or single copies 
of 15 per cent is allowed to accredited agencies. Order from the Society at address above. 
Published monthly at Easton, Pa., by the Society of Motion Picture Engineers, Inc. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 

General and Editorial Office, Hotel Pennsylvania, New York 1, N. Y. 

Entered as second-class matter January 15, 1930, at the Post Office at Easton, 

Pa., under the Act of March 3, 1879. 


Vol 45 NOVEMBER, 1945 No. 5 



Summary. Ansco Color 16-mm motion picture film has been manufactured for 
several years. The bulk of the production has been used by the Armed Forces. In 
field operations, hand methods of developing have been used. However, for first-class 
results an especially designed processing machine is recommended. Such machines 
have been constructed by Ansco and in this paper the author describes the design and 
use of these machines. 

The processing of color motion picture film has, in the past, been 
restricted to a relatively few laboratories specializing in this work. 
With few exceptions, most theater-length productions in color have 
been produced by dye imbibition processes. A similar method has 
been used for the production of most color cartoons. 

Numerous special processes of dye toning, dye bleaching, etc., have 
been proposed and some have been used for the production of short 
subjects. However, none of these has gained wide acceptance, prob- 
ably because of the numerous complications associated with such 
procedures and the unpredictable results obtained by them. 

Dye-coupling color processes, although not new, have come into 
prominence through advancements in organic chemistry which made 
their practical application possible. The Kodachrome process, in- 
troduced some 10 years ago, was the first dye-coupling subtractive 
color process to be carried out on a large commercial scale. It was 
confined for many years to the 16-mm field. Kodachrome developing 
has been carried out only in the laboratory of the manufacturer. 

Agfacolor, another dye-coupling method, was introduced in 1936 
or 1937 and gained some prominence before the war. Although the 
Agfacolor process was not as complicated as previous dye-coupling 

* Presented May 17, 1945, at the Technical Conference in Hollywood. 
** General Manager, Reversal Laboratories, Ansco, Binghamton, N. Y. 




Vol 45, No. 5 

procedures, the developing formulas and procedures were not released 
to the public. 

Ansco has been experimenting with multilayer, dye-coupling sub- 
tractive color procedures for many years. This research has been 
directed toward simplifying the color process to the extent that the 
developing would not have to be carried out by the manufacturer, but 
could be done in any well-equipped laboratory. Our process was 
nearly completed before the war. Through an intensified research 
program carried on because of war necessity, experimental work, 

-Blue sens, 
Green sens. 
Reel 8fens 


FIG. 1. 

which would ordinarily take years, was reduced to months and the 
product was perfected. 

Throughout the war, we have been supplying more and more color 
products for military needs. Some of these products are now also 
being made available to civilian trade in small quantities. While 
some of these products as they become available to the trade seem 
new, they are not new products to us for we have been making many 
of them since the beginning of the war. 

We have been called upon to assist in processing these color prod- 
ucts in all parts of the world and we have gathered a great deal of 
information on the product and its behavior under various processing 
conditions. As these products are converted from war needs to 


civilian needs, this information can be made available for civilian use. 
In addition to this, our laboratories in Binghamton, New York, have 
been processing the material for a number of years on modern proc- 
essing equipment which I am going to describe. 

Ansco Color Film is a multilayer film of the reversible type, pro- 
ducing color by subtractive synthesis. The over-all thickness of the 
film is the same as any ordinary negative film. In normal use, the 
film is exposed in the camera without filters of any kind. During 
exposure, the image is broken up into its color values by the multi- 
layer emulsion. 


Processing Procedure for Ansco Color Film 

Step Time Temperature 

(7) First Develop 12 min 68 F * V* degree 

Short Rinse 5 sec 

(2) Short Stop I 3 min 68 F =*= l / z degree 

(5) Hardener I 3 min 68 F == l / z degree 

Wash and 2nd Expose 3 min 

(4) Color Develop 15 min 68 F l / z degree 
Short Rinse 20 sec 

(5) Short Stop II 3 min 68 F =*= Va degree 

(6) Hardener II 3 min 68 F == V* degree 
Wash 3 min 

(7) Bleach 6 min 68 F * l / 2 degree 
Wash 3 min 68 F V* degree 

(5) Fix 6 min 68 F =*= l / z degree 

Wash 3 min 

(9) Final Wash 6 min 

As shown in Fig. 1, the lower layer of the emulsion records the red- 
dish components of the image, the middle layer records the greenish 
radiations, and the top layer records the bluish colors in the image. 
In this way, a tricolor separation is made automatically and simul- 
taneously in the selectively sensitized emulsion layers at the time of 
exposure. In this respect, Ansco Color Film is entirely conventional. 
The distinguishing feature of the film is the way in which the colors 
are produced. In each of the selectively sensitized emulsion layers a 
colorless dye former is incorporated at the time of manufacture. 
Each layer contains a different dye former which produces color com- 
plementary to the color sensitivity of the layer. 

The top layer is sensitive to blue. After developing, the dye 
former in this layer will contribute yellow to the final picture. The 

316 J. L. FORREST Vol 45, No. 5 

middle layer is green sensitive and after developing, the dye former 
in this layer will contribute the magenta to the final image. The 
lowermost layer of the emulsion structure is sensitive to red and the 


Formulas for Processing Ansco Color Reversible Film 

First Developer H9.8-10 

Metol 3 gm 

Sod. Sulfite 50 gm 

Hydroquinone 6 gm 

Sod. Carbonate 40 gm 

Pot. Bromide 2 gm 

Sod. Thiocyanate 2 gm 

Water to make 1 liter 

Short Stop I and II pH 5.3-6 

Acetic Acid (glacial) 5 cc 

Sod. Acetate 30 gm 

Water to make 1 liter 

Color Developer pH 10-10.3 

Sod. Bisulfite 1 gm 

Colamine 4 gm 

Sod. Carbonate 67 gm 

Pot. Bromide 1 gm 

Water to make 1 liter 

Hardener I and II H 3 . 8-4 . 5 

Pot. Chrome Alum 30 gm 

Water to make 1 liter 

Bleach 0H 6.2 

Pot. Ferricyanide 100 gm 

Pot. Bromide 10 gm 

Dibasic Sod. Phosphate 40 gm 

Sod. Bisulphate 35 gm 

Water to make ' 1 liter 

Fixer PH7.3-8 

Hypo : 200 gm 

Water to make 1 liter 

dye former in this layer will contribute the cyan to the final image. 
The dye formers used in the layers, each producing a different sub- 
tractive color, have the common characteristic that they will each 
combine with the same color-forming developer and remain in their 
respective layers without diffusing. This simplifies the developing 
process and makes it possible to use one color development to produce 


3 different colors simultaneously, rather than an individual color de- 
velopment for each layer. 

The developing process for Ansco Color Reversal Film, given in 
Table 1, consists of 9 essential steps with washes and rinses inter- 
posed where necessary. After camera exposure, the film is given a 
negative development in a metol-hydroquinone developer, Table 2. 
This develops the negative tricolor separation images recorded in the 
emulsion layers. At this point, however, no color is formed because 
color can only be formed when the film is developed in the proper 
color-forming developer. After developing, the film is given a short 
rinse and is short-stopped in an acid short-stop bath to arrest any 
further development. From the acid short-stop bath, the film goes 
directly into a chrome alum hardening bath. After hardening, the 
film is transferred to white light for the remainder of the process. 

A wash follows the hardener and during this wash, the film is ex- 
posed again. Second exposure is accomplished by exposing both the 
back and the front of the film to GE PS-25 Photon 1 ood-type lamps. 
The color temperature of the light is not important. It is important, 
however, to have enough light intensity to expose completely all the 
remaining silver halides in the 3 emulsion layers. 

During the second exposure, all silver halides in the film, which were 
not previously exposed in the camera and which did not develop in 
the negative developer, now become exposed and are ready for de- 
velopment in the color-forming developer. These silver halides 
make up the positive image. In the color-forming developer, this 
image is reduced to metallic silver by the action of the developer. 
Simultaneously with this reduction, reaction products which always 
occur in developing are produced. In a normal black-and-white 
film, these reaction products diffuse through the gelatin and are 
finally washed out in the developer and subsequent baths. However, 
in color film these reaction products combine with the color formers 
in the film and produce dye in proportion to the amount of silver 
halides which were reduced. Thus, as the 3 positive images in the 
film develop to metallic silver, 3 separate dye images are formed in 
situ with the silver images. Concurrently, therefore, a yellow image 
is produced in the top layer, a magenta image is produced in the 
middle layer, and a cyan image is formed in the lower layer. No 
color is formed around the negative image because this image is al- 
ready in the reduced or metallic silver state and, consequently, no 
further reduction can take place. 

318 J. L. FORREST Vol 45, No. 5 

After color developing, all silver halides in all the layers have been 
developed and dye has been formed around the positive image in 
each layer. Because of the silver present in the layers, it is not 
possible at this point to see any color in the film. After color de- 
veloping, the film is rinsed and travels into the second acid short-stop 
bath. This bath arrests any further action of the developer and also 
tends to solubilize the residual color developer remaining in the film 
so that it can be more easily removed in subsequent steps. After the 
second short-stop bath, the film is given a second hardening treat- 
ment in the chrome alum hardening bath. This tends to maintain the 
hardness of the emulsion layer, making it resistant to subsequent 
handling. A wash follows the hardening bath. 

As mentioned previously, the film now contains developed silver 
in all layers. In order to reveal the colors, this developed silver must 
be removed. The film is, therefore, treated in a bleach bath consist- 
ing mainly of potassium ferricyanide and a halogen salt which con- 
verts the developed silver back to silver halides. Since these silver 
halides are soluble in hypo, the film, after a brief wash, is fixed in a 
fixing bath consisting only of hypo. In this bath the rehalogenated 
silver is dissolved, leaving only the composite color images in the film. 
After washing, the film is dried in the conventional way. 

Ansco Color motion picture film can be handled with equipment 
similar to that used for black-and-white film. However, the most 
satisfactory way to develop the film is with a developing machine. 
Hand methods of developing can be used, but such procedures cannot 
be considered production methods and the results obtained with 
them do not compare favorably with the results obtained with ma- 
chine processing. We investigated many procedures for developing 
Ansco Color Film and we were called upon to devise procedures which 
could be used for field operation by our Armed Forces. During these 
investigations, we were not able to find any simple method which 
would produce ideal results. For field operation, however, the Stine- 
man system worked out fairly well and the films processed this way, 
although not first-class from the technical viewpoint, were satisfac- 
tory and provided results where it would have been impossible to 
obtain them by any other procedure. 

All types of Ansco Color Reversible Films are developed by es- 
sentially the same process. However, in mechanizing any process, a 
compromise has to be made between that which it is desirable to do 
chemically and that which it is possible to accomplish mechanically 


and still be consistent with good engineering practice. Developing 
machines of the production type ordinarily are built of a number of 
individual units, each unit carrying a number of film spools. In de- 
signing a developing machine to apply a process having numerous 
steps, it is customary practice to take the minimum time of film 
treatment in any one step and let this time be the governing factor in 
controlling the individual unit size of the machine. For instance, 
in the Ansco Color process having steps requiring several different 
treatment times, such as 12 min, 3 min, and 15 min, respectively, 
the individual unit size is so proportioned that a film treatment time 
of 3 min is given by one unit. The longer times are accomplished by 
increasing the number of units per bath to provide the increased time. 

Our Ansco Color Film developing machines were designed and built 
during the war and in designing these machines a number of rather 
critical production requirements had to be met. The machines had 
to be designed to provide a rather large capacity in order to turn out 
the footage required. Slow-moving developing machines, ordinarily 
used in multistep processes of this kind, could not be considered. 
Furthermore, during the time of critical labor shortage, every labor 
hour had to be used conservatively. It has been our experience that 
it does not require more labor to operate a developing machine with 
fairly high capacity than it does to operate a small one. 

The machine had to handle double-perforated (silent) film, sound 
film, and unperf orated film. It could not place any undue strain on 
the film which might present a breakage hazard. Most of the films 
to be handled on these machines were important originals and 
could not be retaken in case of damage. Furthermore, it was a re- 
quirement that the machines be able to handle damaged film. Some 
of the films exposed in field operation are exposed under very unfavor- 
able conditions which might result in damaged film. In such cases, 
all possible footage must be salvaged. 

In order to meet these requirements, a sprocketless type of ma- 
chine was decided upon and because of our experience with bottom- 
drive developing machines, a bottom-drive mechanism was used. 
Rollers which carry the film by the extreme edges were used through- 
out the machine in order to eliminate the possibility of friction marks 
on the sound track area. Hard-rubber rollers were found to be en- 
tirely satisfactory. The usual commercial hard-rubber film rollers, 
however, caused friction marks on the back of the film and, there- 
fore, especially designed rollers were used. These were procured un- 



Vol 45, No. 5 

bored. It has been our experience that rollers with the center open- 
ings molded are likely to be eccentric. This eccentricity can place 
undue strains on the film as the rollers revolve and for this reason, 
it was found preferable to drill the shaft openings in the unbored 
blanks with precision equipment. 

The developing machines, shown schematically in Fig. 2, which we 
have designed for processing Ansco Color Film, have 25 units in the 
wet section and 4 units in the drying section, making a total of 29 

FIG. 2. 

units which require more than 5000 ft of film for threading. Each 
machine is constructed in 2 parts one part consisting of 7 units is 
installed in the darkroom; the other 18 units of the wet section are 
installed in white light. The roller banks are about 3 l / z ft between 
centers and they are installed 8 in. apart on the top support. These 
are medium-speed production machines and have a variable operating 
range between 40 and 72 ft per min. They are normally operated at 
60 ft per min. The film is driven by friction applied through the 
bottom drive. There are no sprockets. The machine is entirely self- 
compensating for dimensional change in the film. 



The behavior of acetate safety film during the developing cycle is 
rather interesting. There are 2 factors which influence the dimen- 
sional change. One we term the "Humidity Coefficient" and this 
refers to the change of dimension which occurs when the film is 
subjected to moist air or- water. The other factor we term the 
"Temperature Factor." This refers to the change of dimension in the 
film which occurs when it is subjected to changes of temperature 
either in air or in aqueous solutions. In processing, the dimensional 
changes occur continually as the film passes through the developing 
machine. These changes of dimension, seen in Fig. 3, are ordinarily 
not noticeable to the eye, but they are appreciable. Very careful 


FIG. 3 

consideration must be given to them when designing a developing 
machine; otherwise, the film will run with excessive tension in some 
sections and be excessively loose in others. 

In order to compensate for these factors, all top rollers of the 
developing machine are free to move independently of each other. 
To insure this, a stationary separator washer is placed between each 
roller. This permits entirely independent movement of each roller 
without a tendency drag from neighboring rollers. This design will 
permit equalization in the wet section as long as the film is expanding, 
for it permits each loop of film free movement and slippage. It will 
not, however, compensate for shrinkage which must occur in the dry- 
ing operation. In other words, it will take care of the dimensional 
change in the film, as indicated from point A to point B in Fig. 3. 

322 J. L. FORREST Vol 45, No. 5 

From point B to point C a reverse condition exists and the film be- 
comes shorter. In order to compensate for this reversed dimensional 
change, the drying cabinet is underdriven, which means that more 
film is drawn into the cabinet than is taken from it. This difference 
in drive is in relation to the dimensional change which occurs in the 

By designing each section of the machine to compensate inde- 
pendently for these dimensional change factors, the tension on any 
part of the film at any point in the machine never exceeds 30 grams or 
one ounce. With this low strain on the film, breaks in the machine 
are extremely rare. We have some machines in continuous operation 
for more than 6 months at a time without a film break. 

All metallic parts of the developing machines coming in contact 
with the solutions are made of 18-8 stainless steel. The solution 
tanks are made of steel, rubber coated l /s in. inside and l /it> in. on the 
outside. It has been found that coating the tanks with rubber on the 
outside reduces the maintenance cost. 

In operation, the film is fed into an elevator. The elevator holds 
about 120 ft of film. From the elevator the film passes over a break 
signal device into the first developing tank. There are 4 roller banks 
in the first developer giving a total time of 12 min at normal operating 
speed. After developing, the film passes successively into the rinse, 
the short-stop, the hardener, then through the light lock into white 
light. At this point it is given a second exposure. The second ex- 
posure is supplied by 4 GE PS-2o lamps, 2 placed on each side of the 

Simultaneously with the second exposure the film is given a 3-min 
wash and then goes into the color developer. There are 5 banks of 
rollers in the color developer. This gives a total time of 15 min at 
normal operating speed. 

The next 4 units of the developing machine are in single tanks to 
accommodate the steps of rinsing, short-stopping, hardening, and 
washing. After the film has passed through the hardener and its 
subsequent wash, it travels into the bleach. The bleach tank accom- 
modates 2 units and provides 6 min at normal operating speed. 
Bleaching is usually completed in 2 l / z min. The extra time is pro- 
vided to insure thorough bleaching. After bleaching, the film passes 
through a wash tank and then into the fixer. 

The fixer tank contains 2 units allowing 6 min at normal operating 
speed. This insures complete removal of the rehalogenated silver. 


The developing machine is provided with 2 wash tanks following the 
fixer. The first tank contains a single bank of rollers and allows 3 
min of washing. This removes most of the fixer. The second wash 
tank contains 2 banks and provides the final wash of 6 min. The 
residual hypo left in the film after processing falls below 0.005 mg 
per sq in. which is satisfactorily low to insure permanency. After 
washing, the film passes through a double pneumatic squeegee, operat- 
ing at 7 Ib pressure, into the drying cabinet. 

There are 4 banks of rollers in the drying cabinet and drying is 
accomplished in about 15 min. Clean, dry air is supplied to the 
cabinet from a dehydrator. The drying system is entirely closed and 
it is automatic in operation. The temperature is maintained at 86 F 
and the Relative Humidity is held at 35 per cent. In this system, if 
faster or slower drying is desired, the Relative Humidity is adjusted 
at the dehydrator. The air temperature is not changed. We have 
found this method of drying to be very satisfactory from all angles 
and it can be highly recommended. 

The chemical mixing equipment is located on the floor above the 
laboratory and the solutions are fed through rubber pipe lines to the 
processing tanks of the machine. All make-up solutions are main- 
tained at approximately the working temperature. This is accom- 
plished by heat transfer coils located in all storage tanks. The tem- 
peratures of the processing solutions are controlled within narrow 

In order to insure sharp, brilliant color images, the developers, 
bleach, and the hypo are filtered and jetted against the film as it 
passes through the tanks. In the developing tanks, these jets are so 
arranged that the solution is directed against the flow of the film. 
This high turbulence provides a uniform development free from 
streaks usually associated with high-speed machine work. 

During operation, the solutions are continually replenished. The 
amount of replenisher is determined by constant pH control and by 
sensitometry. Sensitometric tests are put through the developing 
machine every 30 min. These strips are checked for speed, color, and 
fog. A pictorial check is also made and in this test, short pieces of 
film, representing 3 different exposures of a predetermined setup, are 
developed together at 30-min intervals. The pictorials consist of a 
color chart and a gray scale. One series is given normal exposure; 
one series is exposed one stop less (this represents underexposure); 
and another series of exposures represents one stop more than 

324 J. L. FORREST Vol 45, No. 5 

normal, which is overexposure. A constant check is made on the 
gray scale for tone and the color patches are checked for general 
color characteristics. From these tests, it is possible to control the 
characteristics of the various solutions so that the quality output 
from day to day is uniform. 

In conclusion, it can be said that Ansco Color motion picture film 
should not be developed by hand processes. Only a developing ma- 
chine will give consistently uniform quality. Machine processing is 
the most satisfactory method of handling the film. Good turbulence 
of the solutions is necessary to insure clear, brilliant, sharp color 
images. No attempt should be made to operate a continuous color 
film developing process without adequate control facilities, for to do 
so will only result in failure. The control procedures necessary with 
machine processing of Ansco Color Film are not difficult to use. 
However, to interpret the result, it is obvious that a thorough knowl- 
edge of each step of the process is required. 


MR. E. E. GRIFFITH: Mr. Forrest, when a sensitometric strip comes off the 
machine, what means do you have for adjusting the relative proportion of the 
color in it? 

MR. FORREST: The relative proportion of the individual color is not adjusted 
during developing because the color proportion is determined during manufacture 
of the film itself, not during the developing operation. 

The color balance is established in reference to developers of known and care- 
fully controlled characteristics. Exhaustion of the developers will affect this color 
balance and it is the purpose of the sensitometric strips to detect this variance 
and thereby provide a guide for the amount of developer replenisher required. 

MR. RALPH ATKINSON : I would like to ask you to go into a little more detail 
as to how you would measure these differences in color in the sensitometric strip. 

MR. FORREST : The color in the strip can be measured with a color densitometer. 
From the density readings obtained from this instrument, the characteristic 
curves of the individual layers can be plotted. 

MR. ATKINSON: Is it possible to vary the contrast of the image at all during 

MR. FORREST: Very little control over the image contrast is possible by de- 
velopment changes. 

MR. A. R. DAVIS: Is it necessary to use four 500-w lamps for the second ex- 
posure, and should this light be of any particular color temperature? 

MR. FORREST: Second exposure requires a considerable amount of light be- 
cause at this stage of the process the film has been slowed down considerably from 
what it was originally at the time of camera exposure. It is not necessary to use 
four 500-w lamps for exposure. At the machine speed at which we operate and 
under our particular conditions, we have found four 500-w lamps to be satisfac- 


tory. At slower machine speeds, a lesser quantity of light could be used. 

In answer to the second part of your question, it is not necessary that the light 
be of any particular color temperature. However, the light should be of a con- 
tinuous spectral quality. The important thing in making the second exposure is 
to provide sufficient light so that all the silver halides remaining in the film are 
thoroughly exposed. There is no danger of overexposing the film at this point. 

MR. CHARLES G. CLARKE: I have noticed that the published directions for de- 
veloping Ansco Color Film are considerably different from the method you have 
discussed for 16-mm film. For example, the washes between the developers and 
the short-stop baths are not called for in the amateur method which has been 

MR. FORREST: The method of developing Ansco Color Film to which you refer 
has been worked out in connection with a developing outfit intended primarily 
for field use under various working conditions, and some of these conditions are 
very unfavorable for developing because of high water temperature. Because of 
this, the washes were eliminated, with a subsequent decrease in the life of the short- 
stop baths after the developers. For a continuous process, such as in developing 
Ansco Color motion picture film, rinses after the developers to prevent contamina- 
tion of the short-stop baths are necessary. Motion picture processing labora- 
tories, however, are usually supplied with cool water. Therefore, this provides 
no hardship in carrying out the continuous procedure. 

MR. S. P. SOLOW: How does the developing machine compensate for dimen- 
sional change of the film ? 

MR. FORREST: In the developing machine described, the film is driven entirely 
by friction. Predetermined overdrive in the first section of the machine com- 
pensates for shrinkage. Elongation of the film takes care of itself automatically 
through the film's disengaging contact with the drying rollers at the points where 
the elongation occurs. Each loop in the developing machine is independent of its 
neighboring loops. This is accomplished by using a stationary shaft for the top 
rollers and providing a washer, which is keyed to the stationary shaft, between 
each roller. This prevents the action of any one roller from influencing the neigh- 
boring rollers and allows each loop 'in the developing machine to adjust itself 
automatically. This is a very important feature of this type of developing ma- 
chine design. In operation, the compensation for dimensional change in the film 
from bath to bath is entirely smooth and automatic. The film travels continu- 
ously and smoothly through the machine. There is no jerking nor is there any 
slack formed at any point. Furthermore, there is no excessive tension built up. 
As mentioned previously, there is less than one ounce pressure on the film at any 
point in the machine. 

MR. H. W. MOYSE: From the illustrations it appears that there is only one jet 
in the developer tanks. Does this provide sufficient turbulation? 

MR. FORREST: Each tank of developer is provided with a series of vertical and 
horizontal jets so positioned that the angle of jetting is against the direction of 
film travel. The developer is supplied to the jets under pressure and this creates 
sufficient turbulence. 

MR. R. C. MARTIN, JR.: Are there any squeegees between the various tanks? 
Isn't there a diluting factor caused by the water entering the various developers? 

MR. FORREST: Rubber squeegees are provided between each tank. Since 

326 J. L. FORREST 

compressed air squeegees atomize the chemicals in the air, they are not used in 
this application. The rubber squeegees are fairly effective in reducing solution 
transfer from tank to tank. 

MR. SOLOW: The developing machine under discussion is a 16-mm machine. 
Do you have any experience with a similar setup for 35-mm? 

MR. FORREST: No, we have not built a 35-mm developing machine of this 
type, although there is no reason why a 35-mm developing machine made in ac- 
cordance with this design would not work equally as well as the 16-mm machine, 
provided appropriate changes were made to accommodate the wider film. 

MR. LYNWOOD DUNN: Why cannot the contrast of the film be changed to a 
considerable degree during developing? 

MR. FORREST: When the recommended developing time for the film is used, 
all 3 layers are developed to the extent that they are in correct balance. This is 
not true when other than recommended developing times are used. Shortening 
or prolonging the developing time will affect certain layers of the film differently 
with the result that the color balance is disturbed. 

MR. DUNN: Would it be possible to restore the color balance by filtering? 

MR. FORREST: If the individual layers were thrown out of balance with each 
other through independent change in the characteristic curves of each, then a 
filter would not restore the color balance. 

MR. DESORM: Could you use fluorescent lamps in place of the "inkies" and 
get enough exposure? 

MR. FORREST: Because of constructional details of our machine, the space is 
too limited to provide for enough fluorescent light to accomplish the second expo- 
sure. We do not recommend fluorescent light for second exposure because of its 
low intensity. There is always the danger that the film will not receive enough 
exposure. If this should occur, the bright mercury lines which are present would 
have a tendency to throw the film out of balance. 

MR. DESORM: Fluorescent lights have been used in black-and-white. 

MR. FORREST: For black-and-white work fluorescent light may be perfectly 
satisfactory because the scale is monotone and is not likely to be thrown out of 
balance by an unbalanced exposure. \ 

MR. MARTIN: How practical is it to divide up the process in other words, 
after the first developer, dry the film and then finish the process on another de- 
veloping machine? Would this procedure cause any loss in quality? 

MR. FORREST: This procedure can be used. It may cause a slight shift in tone 
of the final image. However, this tone shift could be adjusted by making appro- 
priate changes in the rest of the process. If a procedure of this type is used, it is 
very important that the film receive a good wash to remove all residual chemicals 
before it is dried. 

MR. SOLOW: How many of the solutions are circulating hi the machine? 

MR. FORREST: Four solutions are circulated. They are the first developer, 
color developer, bleach, and hypo. The short-stop baths and hardening baths 
are not circulated. 

MR. SOLOW: Do you require any chemical replenishment during the course of 

MR. FORREST: Yes, the chemical solutions are all continuously replenished 
during operation , 


Summary. Practical use has been made by the author of the Eastman mono- 
pack 35-mm color film in photographing the Twentieth Century-Fox Film produc- 
tion "Thunder head." In this full-length feature production every scene was made on 
monopack film including interior sequences, process shots, special effects, and all of 
the exterior scenes. These latter were made under all variations of light conditions, 
and an extremely wide range of background material. 

The paper presents some of the problems encountered, the production techniques 
used, and benefits obtained by "using this method of making motion pictures in color. 

Having recently had a feature production released which was made 
entirely on monopack film, I have been asked to prepare a paper upon 
some of the experiences encountered and the technique attempted. 
The production is Twentieth Century-Fox Film Corporation's 
Thunderhead Son of Flicka, which is the first 100 per cent feature 
production to be made by this Technicolor single-film method. By its 
nature, the requirements of this production gave monopack film a 
thorough test, for a great variety of scenic locations were utilized 
and every conceivable lighting condition was encountered which 
offered a challenge to the color camera. 

Needless to say, much is yet to be learned about monopack, and no 
one at the present stage of development can presume to make any 
statements with final authority. As the monopack process is semi- 
experimental and little information about it is available, my ex- 
periences may be of interest and value to those who contemplate 
using this method of color motion picture photography in the future. 

From the processing standpoint, monopack is more complicated 
than the older methods. Briefly, from the original monopack, which 
is a positive color print, 3 black-and-white separation negatives are 
made. From each of these a printing matrix is made which is then 
saturated with the proper liquid dyes, then transferred and super- 
imposed one upon the other upon clear gelatine-coated film. The re- 
sult is a color print from the original. The Technicolor Motion 

* Presented May 17, 1945, at the Technical Conference in Hollywood. 
** Director of Photography, Twentieth Century-Fox Film Corporation, 
Beverly Hills, Calif, 


328 C. G. CLARKE Vol 45, No. 5 

Picture Corporation has done heroic work in perfecting the process, 
so simply stated above, and they are constantly improving it. In 
the course of making Thunderhead I have seen the color reproduction 
improve, the emulsion speed increase 25 per cent, and the test pilot 
scenes come back to us much nearer normal as to color ratio and 
density. All this despite the fact that Technicolor is processing film in 
greater quantity than ever before and has contributed much of its 
highly trained and experienced personnel to the war effort. With 
these handicaps, plus the lack of needed machinery, Technicolor is 
experimenting with the monopack method, and research and im- 
provements are constantly nearing perfection. 

With these constant improvements that are being made, the 
problems and working methods qf which I speak today may not apply 
in the future, and indeed may have been outdated by the time this 
discussion is published. 

Basically, the technique of photographing in cojor will always 
differ from that of black and white, whatever the method. Color 
pictures have a roundness and third-dimensional effect of themselves, 
which has been striven for in black and white by utilization of light- 
ing methods and employment of certain contrast values. While I 
am not of the schoql which maintains that color pictures must be made 
with flat light, and that pictorial color effects are to be avoided, still, 
I might suggest to the new worker in monopack that it is wise to pro- 
ceed with caution. The tendencies of monopack to intensify con- 
trasts must always be borne in mind. For example, slightly diffused 
daylight reproduces as bright sunlight in monopack, and light blue 
skies become vivid blue. Cloud-filled skies are most ideal, for not 
only do the clouds in the scene add beauty, but those behind the 
camera are reflecting white light into the shadows of the scene being 
photographed and are hence reducing the contrast of direct sunlight. 
Where these conditions do not prevail, reflectors and booster lights 
are used to fill in shadows. Arc lights equipped with ultraviolet Yl 
absorbing niters are used on booster lights as well as those used for 
interior lighting. Bunting diffusers are useful to soften overhead 

Monopack, in common with other color films, has a tendency to 
overemphasize the blue portion of the spectrum. To overcome this, 
an ultraviolet absorbing, almost colorless gelatine filter is used when 
photographing all scenes, exterior or interior. The filter is known as 
WrattenNo. 114-A. 


Those having experience in photographing with Eastman Koda- 
chrome reversal film will find that experience valuable when exposing 
monopack. In many ways the film characteristics of the two are the 

It may be of interest to know that the exposure stops on Thunder - 
head ranged from //2.8 to //16. The interior scenes were photo- 
graphed at//2.8. Exterior exposures varied from //2. 8 for the dark 
canyon and waterfall scenes, which were made under the weakest of 
lighting conditions, to an average of f/8 for the regular landscape 
scenes. Cloud scenes and under-cranking to speed action accounted 
for the smaller stops mentioned. The exterior night scenes were, of 
course, taken in the daytime, and for these scenes the film was one stop 
underexposed to increase density of shadows. In the printing of these 
scenes we requested that the prints favor a bluish cast for the cool 
effect and to simulate the generally accepted conception of moon- 

I have been frequently questioned about the scene containing the 
moon. This scene was made by employing an old trick of cameramen 
that of double-expqsing the real moon over an undertimed scene 
previously made in the daytime. The scene in this case was taken with 
a 35-mm lens exposed at*//ll and the moon was photographed that 
night with a 4-in. focus lens at//2.8 on the same film. 

Many of the same trick devices that are helpful in obtaining scenes 
with black and white may be made with single-film color. For ex- 
ample, in a scene preceding the horse fight, a split-screen double ex- 
posure was made on the location where the wild horse is apparently 
pouncing upon Roddy MacDowell. In the racing sequence there are 
2 rear projection scenes, the plate being also photographed with mono- 
pack. Because of the fine definition of the process plate, these 
"process" shots are not inferior in quality with the normal scenes 
and can scarcely be detected. 

To me, an interesting problem occurred in the sunrise sequence. 
To convey time lapse and distance covered, as well as an opportunity 
for pictorial effect, we dissolved from the night waterfall scenes to a 
sunrise scene. The next series of scenes which followed had to go on 
with the story and be delineated in full light. The sunrise scenes are 
naturally yellow in hue, as should be, but the next cut into the 
"hidden canyon," by necessity, had to be photographed at noon. At 
midday normal sunlight is white and vastly different in color from 
that at sunrise. Correction was made in color printing for the 

330 C. G. CLARKE Vol 45, No. 5 

canyon scenes and we photographed these in such manner that this 
correction could be done without distorting the color of other known 

For the horse fight scenes which followed the hidden canyon scenes, 
we selected a setting which was dominated by orange-red rocks. 
This orange-red background permitted normal color ratio to be re- 
sumed in printing. Thus the color transition was from a very yellow 
sunrise effect into normal daylight within a few feet of film, with only 
2 short scenes color-manipulated. I do not believe audiences are con- 
scious of this time lapse and color transition, but in any event it was 
one that required planning by the color cameraman, and needed the 
cooperation of the director and producer so that the scenes could be 
executed in this manner. In addition, the psychological effect of a 
primitive horse fight against a setting of vivid red background is in 
evidence here. 

In a picture like Thunderhead panoramas were often necessary with 
the action going from front light to direct back light, and vice versa. 
This always presents a problem for color films, for among other 
problems there is the one of the change of hue in the sky between these 
extremes. Under clear conditions the sky in front light is quite blue, 
while the same sky in back light is white. In a motion picture these 
extremes cut together in rapid succession present a variety of hues to 
the audience and appear inconsistent. In rapidly "panning" shots 
there is not much a cameraman can do, though in slower panning 
shots some control is had by using graduated neutral density sky 
filters which are moved into maximum absorbing position as the 
camera pans into the strong back light. The exposure variable is 
controlled by the dissolving camera shutter or by the lens diaphragm 
during the taking of the scene. 

Back light has been considered taboo in color, and in monopack it 
is difficult because of the added contrast. I must confess here to hav- 
ing put the film to severe tests in many cases. The porcupine se- 
quence was deliberately photographed in direct back light because we 
wished to convey in this series of scenes the idyllic impression that 
highlights and reflections on running water and waving grass can 
give. Of course, a white horse and a lot of reflectors helped. 

It may be proper here to enumerate some of the disadvantages of 
monopack in its present state of development. 

First, there is the disadvantage in obtaining rush prints within the 
length of time we have considered reasonable with regular Technicolor 


and black-and-white prints. Naturally, to those using a new and 
unfamiliar medium, a print of maximum quality is desirable as soon 
as possible in order that a visual check may be made on the work that 
is being done. In this way errors of exposure and lighting may be cor- 
rected and future like mistakes prevented. With any reversible film, 
exposure must be correct within very narrow limits, as exposure de- 
termines photographic quality and correct color rendition. It is a 
tribute to the progress made by Technicolor that the exposure lati- 
tude has been widened with monopack and a certain amount of over- 
or underexposure may be controlled and brought into line through 
the latitude of the release print process. 

At present, we view our "rushes" from a black-and-white print 
made by the reversal method. From an economic standpoint this is a 
great saving of film, for 6 more films would be necessary 3 separa- 
tion negatives and 3 matrices for each scene printed in color. After 
a little experience in judging these reversal prints, the cameraman can 
gain quite a good conception of what quality the final color print will 
be. The film editor is at a disadvantage when cutting a production 
from these prints, for better selection and matching could be done if 
he were cutting from a color print. Rather than the black-and-white 
print, it would be better for the production if a contact print could 
be made on monopack, or some type of integral tripack film for the 
working print. The added cost of this film would be largely offset 
by the elimination of the color "pilot" test scenes which are now 
furnished, but which would no longer be needed. 

Secondly, the film processing at the present time seems to add con- 
trast, hence there is some exaggeration of color. In exterior scenes 
where the spectator has no standard values for comparison, this in- 
tensification is not too objectionable, but in the rendition of skin 
textures in close-ups, the audience knows what values to expect and 
top great a deviation from these norms results in unacceptable color 
rendition. As interior scenes are mostly composed of close shots of 
people, these scenes require a soft lighting technique for the most 
satisfactory results. 

The question might be asked, "Why use monopack while it is in its 
experimental stages when the Technicolor 3-strip method is pro- 
ducing such fine results?" My answer would be, because of the 
greater facility in using standard motion picture cameras as against 
the heavier and more cumbersome 3-strip cameras. This is a most 
desirable requirement when photographing scenes in rough and 

332 C. G. CLARKE 

difficult locations. In this connection, it may be stated that many 
of the scenes in Thunderhead could not have been put on the screen 
had it not been for the portable cameras that were used. Not only is 
production time saved in placing cameras, but camera angles can be 
employed and a greater selection of photographic lenses may be 
utilized. This is particularly so with the shorter focal length lenses, 
commonly called wide-angle lenses. Modern pictures utilize the ad- 
vantages of wide-angle lenses for their greater depth of focus and 
interesting perspective renditions. With monopack it is possible 
to employ all the advances of the black-and-white photographic 
technique plus the greatly enhanced value of a production in full 

Technicolor's monopack method provides producers with a medium 
for making productions in color when 3-strip color cameras are not 

Prints from monopack reveal a great improvement in definition. 
To me, this is extremely important, for many film patrons complain 
"that they like color pictures but they hurt their eyes." I feel that 
it is not the color that has troubled them but rather that their eyes be- 
come strained and fatigued from trying to sharpen the diffused color 
print shown on the screen. While monopack prints leave more to be 
desired in this respect, still, they are sharper and better defined than 
the other methods now being shown. Separation negatives obtained 
from the monopack original are all equally sharp. 

In this connection, all the lack o^ definition cannot be laid at the 
feet of color films. There is an appalling lack of care to focus the films 
properly in the theater's projection. Black-and-white film requires 
one setting and the thicker color film another in order to properly 
focus it. With the projectionist constantly interchanging films, both 
suffer neglect. It is to be hoped that audiences can hold on until the 
day when all pictures will be in color. 

While I have mentioned the bad as well as the good things about 
monopack, I feel on the whole that monopack has given a good ac- 
count of itself. When one realizes the enormous sums of money and 
the years of research that have been put forth by scores of cqlor firms 
attempting to produce a successful product, I feel that the com- 
paratively new monopack is off to an excellent start. 

It is hoped that the suggestions I have tried to make, together with 
the results that are obvious when viewing our monopack picture and 
others to come, will aid users of this color method to obtain the wholly 
satisfactory results that are inherent in it. 


Summary. This paper describes an assembly of 5 equalizers arranged so that 
they can be controlled by one hand, thereby leaving the other hand free for dialogue level 
adjustments. The equalizers are capable of lowering or raising the response in 5 
different frequency bands without creating changes in reproduced level. 

It is well known that it is necessary to equalize the frequency char- 
acteristic of recording systems when recording dialogue for motion 
pictures. Low frequencies are usually reduced, the shape and amount 
of the low-frequency reduction depending upon the microphone re- 
sponse, the closeness of the microphone to the actor, the low-fre- 
quency reverberation of the set in which a scene is being recorded, the 
effort with which the dialogue is read, the amount of bass in the 
voices, the loudness at which the dialogue will be reproduced in 
theaters, the frequency characteristics and reverberation character- 
istics in theaters, and other reasons. Also, to obtain best results it is 
often necessary to raise the high frequencies, or the mid-high fre- 
quencies because of theater high-frequency characteristics, micro- 
phone characteristics, or because of the position of microphones rela- 
tive to the persons speaking. 

A fixed equalization characteristic can be used for correcting fixed 
characteristics, but variable equalization is needed to correct unavoid- 
able variations which occur from microphone positions, voice effort, 
voice characteristics, and set acoustics. In the past, efforts have been 
made to take care of all of the variations as they occur, during the 
original recording, but the tendency now is to apply a fixed amount of 
correction when making the original dialogue recording, and to vary 
the equalization during rerecording. One reason for this change is 
that the horn system, used in rerecording, offers a better facility for 
judging frequency response than the headphones used in making the 
original recording. 

* Presented May 15, 1945, at the Technical Conference in Hollywood. 
** Sound Dept., Paramount Pictures, Inc., Hollywood. 




Vol 45, No. 5 

During the last few years a great deal of care has been taken in 
making the corrections during rerecording. Multistage equalizers 
have been built for correcting the low frequencies, the mid-high fre- 
quencies, and the high frequencies. Recently Paramount has been 
using additional equalizers for controlling the shape of the low-fre- 
quency equalization, and for raising or lowering the range between 
300 cycles and 800 cycles. The number of controls for operating the 
equalizers has increased to five, and it has become a difficult problem 

FIG. 1. Control levers. 

to handle them properly, even though one man has been assigned the 
sole task of controlling the dialogue volume and equalization. Proper 
control of the dialogue level requires almost continual adjustment of a 
mixer dial, and consequently it is necessary to operate the 5 equalizer 
controls with one hand. To facilitate this, a 5-unit compensated 
equalizer assembly having lever controls has been built. 

The assembly consists of 5 bridged- T constant resistance equalizers 
mounted in the mixing console, and electrically connected to control 
attenuators and compensating attenuators, which in turn are me- 
chanically connected to a group of control levers through dial-tuning 



pulleys and cables. Figs. 1 and 2 show the control levers and at- 
tenuator assembly. Relays are operated by cam switches on the 
attenuator shafts, and are connected so that as the levers are moved 
through the center position, away from the operator, the equalizer 
elements are switched from connections which "lower" the frequency 
response to connections which "raise" the frequency response. 

FIG. 2. Attenuator assembly. 

Fig. 3 shows the complete electrical circuit, except that the relay 
circuit has been omitted for simplicity. To show the way the equal- 
zer elements are switched, the first 3 equalizers have been shown 
with the "lower" connection, and the last 2 equalizers are shown 
with the "raise" connection. The 5 equalizers are connected in tan- 
dem, and are followed by the compensating attenuators and an am- 
plifier having 35 db gain to give a zero loss system. The control 
attenuators have 8 steps for lowering the frequency response 1, 2, 3, 4, 
6, 8, 11, or 15 db, and 7 steps for raising the frequency response from 
1 to 7 db in steps of 1 db. The compensating attenuators have 7 db 



Vol 45, No. 5 

loss on the "off" position, and for all steps where the response is being 
lowered, and decrease in 1-db steps from 7 to db when the con- 
trol pots are raising the response from to 7 db, respectively. 

The equalization characteristics are shown in Fig. 4. The high- 
pass filter normally connected in the dialogue circuit is also shown. 
The full lines indicate the range in which each equalizer is normally 

A wide variation in low-frequency shaping has been found to be 
desirable, and this is accomplished by using the relatively sharp 90- 
cycle equalizer in combination with the relatively flat nonresonant 
low-frequency equalizer. These equalizers are used mainly to obtain 


FIG. 3. Equalizer circuit. 

a general reduction in "bassiness" and for correction of individual 
scenes. The original recording is normally recorded with a slight 
excess of bass to avoid getting "thin" recordings on raised voices. 
When it is necessary to add low frequencies it is usually done with the 
90-cycle equalizer. 

The 500-cycle equalizer is usually used for a general reduction in 
mid-low frequency "heaviness," or for giving body or fullness to 
scenes which are inclined to be thin. It is seldom necessary to use 
more than one or 2 steps of this equalizer. 

The 3500-cycle equalizer is used extensively for treating "dull" 
scenes, and for improving intelligibility. It is very effective in re- 
storing sufficient high frequencies for good intelligibility without 
causing an undesirable increase in film background noise. 



The high-frequency nonresonant equalizer is too broad to use in 
large amounts, but is very effective in increasing "brightness" when 
used on steps +1 to +3. 


000 ZOOO ^000 SOOO 

LOW FREO. CQ.(NOM.c*Mrrr) 


100 200 


4O 100 200 50O 1000 3000 


FIG. 4. Equalization characteristics. 

In practice it is customary to set the controls to give the most de- 
sirable frequency response for the particular picture being rerecorded, 
and to deviate from these positions as required. Single dialogue lines 
which have been poorly recorded can be equalized so that fairly uni- 
form intelligibility and frequency response are obtained throughout 
the picture. 


MILLER, W. C., AND KIMBALL, H. R.: "A Rerecording Console, Associated 
Circuits, and Constant B Equalizers," /. Soc. Mot. Pict. Eng. t 43, 3 (Sept., 1944), 
p. 187. 

338 W. L. THAYER 

Sound Engineering," D. Van Nostrand & Co. (New York), 1938, p. 228. 

ZOBEL, O. J.: "Distortion Correction in Electrical Circuits with Constant 
Resistance Recurrent Networks," Bell. Syst. Tech. /., VII (July, 1928), p. 438. 

FLETCHER, H. W.: "Speech and Hearing," D. Van Nostrand & Co. (New 



Summary. This paper gives a description of a new 2 -way loudspeaker for 
theaters. New permanent magnet low-frequency and high-frequency units having 
replaceable diaphragms are described. These units are combined in a horn system 
having the following advantages: A higher efficiency, extended frequency range, per- 
manent magnet units providing higher air gap flux densities, elimination of back- 
stage radiation from the diaphragms, better transient response, and an improved 
over-all presence. 

The use of the. present 2-way multicellular horn systems over the 
period of the last 10 years has permitted the theater to give the 
public a sound quality representative of the sound recording tech- 
nique available during the same period. However, during this 10- 
year period, experience has been gained indicating that still better 
recording technique is possible when better loudspeakers are avail- 
able for monitoring purposes. 

Accordingly, new loudspeakers are now being designed for this pur- 
pose, and it is the hope that they will bring the quality of sound even 
nearer to the ideal objective of sound engineers. We are sure that all 
of us will agree on this objective; we want improvement in both 
high- and low-frequency units, and we want the use of these units in a 
loudspeaker system having greater efficiency, higher power capacity 
per unit, better transient performance, an extension of the frequency 
range, a higher definition in quality, and a better over-all presence. 

But it is necessary to have a new motion picture loudspeaker sys- 
tem (Fig. 1) in order to gain these improvements. Before such a new 
horn system could be developed, other things had to come. We had 
to have new methods of manufacturing diaphragms, we needed better 
voice coil construction, and magnets had to be developed that would 
be considerably superior to anything we have had in the past and 

* Presented May 14, 1945, at the Technical Conference in Hollywood. 
** Altec Lansing Corporation, Hollywood. 


340 J. B. LANSING AND J. K. HILLIARD Vol 45, No. 5 

these magnets came along as one of the developments in the war 

In the past, poor presence has been one of the principal deficiencies, 
which can be attributed to several causes. As an example, dips in the 
250-500-cycle region tend to give the effect of individual low- and 
high-frequency sources. Resonances in the low-frequency units 
and horns have accentuated narrow bands. Backstage resonance, 

FIG. 1. Front view of A-2 loudspeaker. 

caused in part by radiation from the rear of the speaker system, 
has caused detrimental hangover and masking of the auditorium 
sound with an attendant loss of presence. Long air column low-fre- 
quency horns become involved in phasing trouble and loss of presence 
is encountered owing to the fact that the apparent source of sound 
tends to recede back in the horn progressively with an increase of 
frequency. Folding of the horn tends to limit the frequency range in 
proportion to the sharpness and number of the turns. Rigidity is 
necessary so that the walls of the horn will not vibrate and dissipate 


sound power by absorption and also give uncontrolled directional 

With this long list of difficulties in mind to begin with, new low- 
and high-frequency units were designed. These units have improved 
impedance characteristics, 
longer life diaphragms which 
dissipate more power safely, 
permanent magnet units with 
diaphragms that can be changed 
easily, and new magnetic cir- 
cuits combining long life at high 


One of the basic improve- 
ments in the loudspeaker sys- 
tem has resulted from the de- 
sign of a new high-frequency 
unit. The larger metallic dia- 
phragm units available in the 
past have used the annular 
type of compliance. This type 
of compliance, while adequate 
at high frequencies, did not 
^provide the necessary ampli- 
tude at lower frequencies. As 
a result, both the power and 
frequency characteristics in the 
region from 250-500 cycles 
have been found inadequate 
owing to the inability to handle 
the necessarily large excursion 

Back in 1925, E. C. Wente 1 of Western Electric recognized the 
necessity for a tangential compliance in loudspeaker units and micro- 
phones in order that the distortion be held to a minimum at low fre- 
quencies. Diaphragms used in the Western Electric 555 receiver had 
such a compliance. 

Recently, J. B. Lansing has developed a hydraulic method of 
drawing metal diaphragms which simplifies the process considerably 

FIG. 2. Front and rear views of 288 re- 
placeable diaphragm assembly. 



over the previous methods. As a result, it is now possible to draw 
larger diaphragms and provide a tangential compliance in these 
larger sizes. This method provides an amplitude approximately 3 
times as great as the annular type and the increased length of com- 
pliance insures that the diaphragm (Fig. 2) can operate at the re- 
quired amplitude without undue strain. 

The voice coil is wound with rectangular aluminum ribbon which 
has been treated with a temperature resistant varnish so that it will 
safely dissipate higher power without damage. The use of edgewise 
wound ribbon provides more volume of conductor in the magnetic cir- 
cuit which, in turn, increases 
the efficiency. Beryllium cop- 
per leads are spot-welded to 
the voice coil wires. This pro- 
vides a heavy duty lead (Fig. 
2) which will not fatigue under 

The entire voice coil and dia- 
phragm assembly is mounted 
in a cast bakelite ring. The 
voice coil leads are clamped 
and soldered under flat termi- 
nals and a screw is provided 
for fastening each connecting 
lead to its binding post. By 
removing the leads to the bind- 
ing post and 6 screws which 
anchor the bakelite ring to the top plate of the field assembly, the 
diaphragm and voice coil may be removed for replacement purposes. 
Two dowel pins are provided for alignment. The use of this 
method of mounting the diaphragm assembly permits its removal 
even though the magnet is charged. As a result, field replacement 
of the assembly is a simple operation and does not require that the 
entire unit be returned to the factory nor does it require any special 

The entire unit (Fig. 3) weighs 21 Ib which is considerably lighter 
than previous units of comparable efficiency and power capacity. 

The impedance of the unit when mounted in a properly matched 
horn is approximately 24 ohms over a wide frequency range. 

Excitation of this new high-frequency unit is obtained from a 

FIG. 3. 

Side view of 288 high-frequency 


newly developed Alnico No. 5 permanent magnet material. The flux 
density is greater than has been used in the best separately excited 
units supplied. 

The magnet itself is of the center core type. The soft magnetic 
material forming the path between the pole pieces is amply designed 
so that the flux is conducted through the outside walls and up to the 
air gap with little loss. The external leakage loss is extremely low in 

FIG. 4. View of 515 low-frequency loudspeaker. 

this design and as a result does not attract metal objects in the im- 
mediate vicinity. The efficiency of the 288 high-frequency unit 
when mounted in a suitable multicellular horn is such that a sound 
level of 98 db (ref. 10 ~ 16 w per sq c) is produced at 5 ft distance for an 
electrical input of 0. 1 w at 1000 cycles. 


The 515 low-frequency unit is mounted in a 15-in. die-cast frame 
which assures permanent alignment of the cone and voice coil assem- 



bly as shown in Fig. 4. It uses a seamless moulded cone having an 
effective area of 123 sq in. and is moisture resistant. An edgewise 
wound copper ribbon coil (see Fig. 5) is attached to the cone and a 
dome is inserted in the center of the cone to provide the maximum 

active vibrating area. The use 
of edgewise wound copper rib- 
bon improves the space factor 
over that of round wire and 
since more conductor material 
can be placed in the air gap, 
the efficiency is raised and the 
operating temperature de- 
creased. Since the 3-in. voice 
coil diameter is considerably 
larger than the 2- and 2 1 /2-hi. 

(diameter coils formerly used, it 
has a correspondingly increased 
ability to handle higher power 
without undue temperature rise, 
and, as a result, the efficiency is 
H little affected with changes in 


A clamping ring fastens the 
outer rim of the cone to the 
frame. The inner spider as- 
sembly is held down by means 
of screws so that it is a simple 
operation to remove the entire 
voice coil and cone assembly 
for replacement purposes. 

An Alnico No. 5 permanent 
magnet is provided for the field 
excitation. The total energy 
available with this magnet is 
greater than that previously supplied in energized units now being 

The resonance of the cone and voice coil assembly is 40 cycles in 
free air. The impedance of the unit is approximately 20 ohms as 
normally used. The unit will safely handle an input signal of 25 w. 
The unit is 15 3 /i 6 in. outside diameter, 8 in. deep and weighs 33 Ib. 

FIG. 5. Replacement cone and voice 
coil assembly for 515 low-frequency 



The N-500-C dividing network used (see Fig. 6) is a parallel- type 
constant resistance network. It consists essentially of a low- and 
high-pass filter designed to operate from a common source at their in- 
put ends. The insertion loss of the network is less than 1 / 2 db. The 
crossover point is at 500 cycles and at this point the power is divided 
between the high- and low-frequency legs such that each branch is 
down 3 db. The attenuation slope is approximately 12 db per octave 
on either side of the crossover frequency. 

FIG. 6. N-500-C dividing network. 

Provision is made for 5 steps (1 db each) of attenuation in the high- 
frequency output. This is accomplished by changing the shorting 
strip held under 3 screws. 

The input impedance of the dividing network is 12 ohms. 


The new improved low-frequency horn (see Fig. 7) which is used 
for medium size theaters has two 515 low-frequency units mounted 
beside each other in a straight exponential horn. The area of the 
throat of the horn has been made approximately equal to the area of 
the 2 diaphragms, giving a loading factor of unity. This increased 



loading over that formerly used provides better damping of the units 
and increases the excursion of the diaphragm. 

These new units are enclosed from the rear (see Fig. 8) so that radia- 
tion from the back side is dissipated in the enclosure. However, at 
frequencies below 100 cycles this dissipation is not complete and ports 
are provided in the front of the speaker, below the mouth of the horn. 
These ports provide an acoustic impedance which raises the output 

FIG. 7. Front view of A-4 loudspeaker system. 

several decibels around 50 cycles. Wings are provided for additional 
low-frequency loading. 

One 288 high-frequency unit is used on the proper horn which de- 
pends upon the shape of the room. 

Early experience with the first 2- way loudspeakers indicated that 
the relative phasing of the 2 units was important. 2 For correct 
phasing the 2 horns must have equal path lengths. The design of 
previous loudspeaker systems has not permitted this optimum phas- 
ing condition to be obtained. Measurements recently made out of 
doors in free space indicate that wide variations in response can be ob- 


tained at the crossover frequency when the horns are shifted so that 
the mouths of the horns are not in the same vertical plane. This new 
horn system has a path length such that the tip of the high-frequency 
multicellular horn mouth is exactly in line with the mouth of the new 
low-frequency horn for correct phasing, and under these conditions 
there is no variation in the response at the crossover. 

FIG. 8. Rear view of A -4 loudspeaker system. 

The A-4 medium size horn system (see Fig. 7) has a rated capacity 
of 40 w. Destructive tests indicate that this rating provides a safety 
factor of greater than four over that necessary to damage the unit. 

The A -2 large size horn system (see Fig. 1) is composed of 2 low- 
frequency horns placed side by side and two 288 high-frequency 
units mounted on a double throat. The dividing network is mounted 

348 J. B. LANSING AND J. K. HILLIARD Vol 45, No. 5 

on the side of the baffle. The installation time is materially decreased, 
since the only wires needed are those from the output of the amplifier. 

Sufficient damping of the vibrating elements of the units are pro- 
vided in the magnetic circuit so that it is not necessary to provide 
additional damping from the driving amplifiers. In the past it has 
been customary to adjust the amplifier output impedance to a value 
of approximately one-half to one-third of the average loudspeaker 
impedance. Improved performance can be obtained with the new 
loudspeaker when the amplifier and loudspeaker impedances are 
approximately equal. 

Anticipating that these new systems may be called upon to provide 
the sound channel in television work, it was necessary to restrict the 
stray magnetic field in order to prevent magnetic distortion of the 
television image caused by the proximity of the cathode-ray tube. 
Additional benefits from these features of the design are increased 
efficiencies owing to lower magnetic losses, and the fact that it is 
possible for these new permanent magnet units to be handled with- 
out endangering wrist watches or other devices which may be sus- 
ceptible to damage from magnetization. 

The efficiency of this new horn system is approximately the same as 
Dr. Fletcher's system 3 and is from 2-8 db higher than commercial 
loudspeaker systems now in use in theaters. 

An A-2 Altec Lansing horn system is now installed and is being 
used at the Pantages Theatre in Hollywood where field tests are being 
conducted. The Academy Research Council Standards Committee 
has recently had a meeting at the theater and has listened to the 
Academy test reel and current studio release product. 

The electrical characteristic tentatively selected by this group is 
identical to the published metallic diaphragm characteristic 4 which 
has been adopted by the Committee for the range above 300 cycles. 

Since the new horn systems have a smoother low-frequency re- 
sponse, experience to date indicates that a bass boost as much as 2 
db at 50 cycles may be used with present product without interfering 
with dialogue quality. The straight low-frequency horn provides an 
unattenuated output up to and beyond the 500-cycle crossover point. 
This increased output in the region from 300-500 cycles over that of 
older horn systems adds materially to the presence and loudness of 
the over-all system. Recording and rerecording staffs in the studios 
indicate from their listening tests that over a period of time it should 
be possible to fully utilize the increased performance of the new loud- 


speaker system so that a smoother and more extended frequency and 
volume range can be reproduced. 

It is our feeling that the early presentation of these loudspeaker 
systems will be a distinct aid to the sound equipment manufacturers 
in preparing their designs of future theater sound systems in order 
that the industry may not be limited to the quality standards es- 
tablished by older loudspeaker systems. 

Similarly, the higher quality standards which can be reached 
through the use of these loudspeaker systems influence studio re- 
cording and monitoring practices. Because of the long interval which 
necessarily intervenes between the recording of a motion picture and 
its presentation to the public, considerable time must necessarily 
elapse before the full influence of the advancements in recording and 
reproducing can be presented to theater patrons. 

The advantages of the new Altec Lansing loudspeaker systems are 
summarized as follows : 

(1 ) Higher efficiency, 

(2) Wider frequency range with a better transient response, 
(5) New permanent magnets, 

(4} Diaphragms that are easily replaceable, 

(5) No backstage resonance, 

(6) A higher safety factor at increased power, 

(7) An improved over-all presence with a much better definition of sound 


1 WENTE, E. C., AND THURAS, A. L.: "A High Efficiency Receiver for Horn- 
Type Loudspeaker of Large Power Capacity," Bell Syst. Tech. J., VII, 1 (Jan,, 
1928), p. 140. 

Engineering," D. Van Nostrand & Co. (New York), 1938, p. 109. 

3 WENTE, E. C., AND THURAS, A. L.: "Loudspeakers and Microphones," 
Bell Syst. Tech. J., XHI, 2 (Apr., 1934), p. 259. 

"Revised Standard Electrical Characteristics for Two- Way Reproducing Systems 
in Theaters," Tech. Bull. (Oct. 10. 1938). 


Summary. The purpose of this paper is to discuss briefly the method of using 
so-called reverberation chambers for recording. After a general consideration of 
decay of sound in reverberant rooms, there is discussed the case where the reverbera- 
tion of the chamber is superimposed upon that of the room in which the sound was 
originally recorded. Also discussed in the paper is a brief description dealing with 
the construction of a double reverberation chamber. 

In the recording of sound on film or wax it is frequently desirable to 
add a reverberatory quality to the recording after its completion. 
This may be accomplished by reproducing the sound in a highly rever- 
berant room the so-called reverberation chamber and "mixing" 
the output from a microphone in this room with the original record- 
ing in a process known as "dubbing" or rerecording. 

Surprisingly, when the electrical level of the reverberated signal 
is as much as 20 db below the electrical level of the original recording 
at the mixing panel, the combined reproduced signal conveys a 
strong impression of reverberation in every syllable of speech, and 
chord or passage of music. 

Unlike in other means, electrical or mechanical, of adding a rever- 
beratory note to a recording, the chamber method provides both the 
proper growth characteristic and the decay quality of sound in a live 
enclosure. Delay networks, magnetic tape recordings, and other 
devices for achieving synthetic reverberation usually permit only 
provision for the decay characteristic; no attempt is made to intro- 
duce the growth characteristic, since the latter is less essential in an 
approach to total reverberation. 

It is interesting to plot the growth curves of sound and the corre- 
sponding decay characteristics for a number of rooms. Since we are 
dealing with enclosures having a little absorption, the following 
equations may be used to describe the build-up and the "die-away" 
process of sound in a confined space. 

* Presented May 15, 1945, at the Technical Conference in Hollywood. 
** RCA Victor Division, Radio Corporation of America, Hollywood. 



Growth : 

E = 1 


10 log, -^~ = 10 Iog 10 

77 o 

E = -ut 



^ 10 iog 10 -gr = ~Y 

where = energy-density at time, 

Eo = steady state energy-density 
T = reverberation period 

Figs. 1 and 2 show the sound-growth and sound-decay curves for 
rooms that have different reverberation times, Fig. 1 being plotted 
on a percentage basis, while Fig. 2 employs the decibel as the unit 
for the ordinate. The curves are plotted on the assumption that the 
power output of the source in the different rooms is such as to pro- 
vide the same value of steady-state energy-density in each enclosure. 
If the rooms were identical in shape, and the power output of the 
sources were the same, then the steady-state energy-density of the 
reflected sound only in the room with the 8-sec reverberation would 
be at least 8 times that in the room with the one-second reverberation. 
This may be calculated from the equation of the reflected sound 

_ 4P (1 - a) 


where P = power output of source 

a = average absorptivity of wall material 
S = total interior surface 



Vol 45, No. 5 

For the addition of reverberation to recordings made on film, a 
room with a reverberation period of approximately 4 sec appears ade- 
quate. A chamber of 4000 cu ft volume, with walls and ceiling made 


FIG. 1. Curves illustrating growth and decay of sound in different rooms, 
plotted on a percentage basis. 

2 3 




FIG. 2. Curves illustrating growth and decay of sound in different rooms, 
using the decibel for the unit of the ordinate. 

of concrete, answers this purpose very well. If the mean dimensions 
for the height, width, and length of the enclosure are 12.5 ft, 16 ft, 
and 20 ft, respectively, the total interior surface comes to approxi- 
mately 1540 sq ft. Crediting concrete, 6 in. thick, with an absorptiv- 
ity of 0.03 sabine at 1000 cycles, the total absorption comes to 46.2 



sabines, and the reverberation time therefore to 4.33 sec. Mean di- 
mensions are indicated because the preferred shape of the enclosure 
is nonrectangular, in order to avoid flutter echoes. 

It is interesting to consider the decay characteristic of the sound 
which actuates the microphone in the chamber. The sound which 
was originally recorded in the recording stage is itself characterized 
by the reverberation of the studio. During reproduction the decay 
characteristic of the chamber is superimposed upon that of the stage. 



1 V 





.10 .12 


FIG. 3. Curves showing sound decay for combined reverberations. 

The equation 1 for the combined reverberation times is given as 
follows : 



14* 14* 

T - T 2 e~ TT 

rig-fT - 
~ r,- 



In the case in which the reverberation time of the recording studio 



Vol 45, No. 5 

is one second and of the reverberation chamber is 4 sec, the equation 
reduces to 

db = 10 log (g-H< -4e-3.5/) -4.76 

The curve is plotted in Fig. 3 together with one representing the 
combined reverberation of 0.5 sec for the recording studio and 4 sec 
for the reverberation chamber. Shown also in Fig. 3 are the theo- 
retical decay characteristics of sound in rooms that have 0.5-sec, one- 
second, 4-sec, and 8-second reverberation. It is seen that for short 


















FIG. 4. Curves showing direct and reflected energy-density as a function of 
distance between source and pickup. 

initial intervals of time the slope of the decay curve of the combined 
reverberation times is rather large. This may indicate that, to the 
ear, the combined reverberation time is in excess of anything that 
may be expected by an arithmetic addition of the individual rever- 
beration periods. Thus, superimposing a reverberation period of 4 
sec upon sound recorded in a room with a reverberation period of 
one second has, for short initial intervals, the effect of sound decay- 
ing in a room of 8 or more sec reverberation. This may account 
for the low electrical level of the reverberated signal necessary (at 
the mixing console) for its combination with the original recording 
to obtain the desired reverberatory character in the rerecording. 
Another reason for the low electrical level required of the rever- 


berated signal is the fact that the microphone represents only one ear. 
In binaural hearing, the ear is to some extent capable of suppressing 
unwanted sound, whether direct or reflected, and to concentrate 
only on the desired sound. This can be readily demonstrated by 

*A : * : ; *% x^ ^-^ : *^:fo -- : ff 





FIG. 5. Double reverberation chamber. 

closing one ear in the reverberation chamber, in which case the re- 
verberation appears considerably prolonged. 

Another reason for maintaining a low electrical level for the rever- 
berated signal is, of course, an attempt to preserve as much as pos- 
sible the intelligibility of the dialogue. The ear is evidently able to 

FIG. 6. Block diagram of rerecording channel using a reverberation chamber. 

judge the reverberation of a room by the audible, slow trailing-off 
of the sound intensity at the end of words. 

In the case of music, where longer reverberation tends to provide a 
richer or more pleasing quality, the electrical level of the reverberated 
signal is kept higher. Still, the definition of instruments can be 
preserved remarkably well by this means. One may indeed con- 
sider whether this type of reproduction does not supply a superior 



Vol 45, No. 5 

rendition, unattainable in any other way, since both clarity of instru- 
ments and an undertone of prolonged decay exist simultaneously. 

Different recordings call for the addition of different amounts of 
reverberation, and sometimes, for the addition of different types, of 
reverberation characteristics. The reason for this is that, in sound- 
on-film recordings, a large number of different sound effects have to 
be included. If the dialogue is recorded in a cellar, tunnel, hull of a 
ship, etc., but the incident sounds (footfalls, jack-hammer drives, 
engine noise) are not included in the original recordings, it is desirable 
to match the character of these effects with the "room- tone" existing 
in the surrounding in which the speech was recorded. Some varia- 
tion in the reverberation characteristic can be effected by changing the 



FIG. 7. Triple reverberation chamber. 

distance between loudspeaker and microphone, since this will change 
the ratio of direct-to-reflected sound at the microphone. Fig. 4 
shows the direct and the reflected energy density for the room in- 
dicated in the figure. 

The character of the reverberated signal may be altered more per- 
ceptibly by dividing the chamber into a small and a large room. Since 
the number of normal modes and their spectral distribution is de- 
cidedly different in 2 such enclosures, a considerable variation in the 
quality of the signals may be effected by using one or the other of the 
2 rooms. If a door is provided in the partition between the 2 cham- 
bers, a further variation will result by placing the speaker in one of the 
rooms and the microphone in the other. Such a door acts like an 
acoustic high-pass filter, thereby making the reverberation char- 
acteristic of each room a function of the door opening. 


A change in the character of the sound picked up in the chamber 
can, of course, also be secured by using different microphones. 

Fig. 5 represents the plan of a dual reverberation chamber, of 
which the walls as well as the ceiling and the floor were kept at a 
slant to avoid echoes. Two such dual chambers employed concrete 
for the material of the walls, the ceiling, and the floor. A massive 
door, 5 ft wide and 6 ft high, weighing approximately 450 Ib, could be 
rotated by means of a knob located at the rerecording console; in this 
manner it was possible to change the reverberation characteristics of 
the rooms easily and quickly. 

Fig. 6 shows a block schematic of a recording channel employing a 
reverberation chamber. 

Fig. 7 illustrates a triple reverberation chamber, designed for the 
purpose of achieving extreme flexibility in rerecording work. 


l HiLL, A. P.: "Combined Reverberation Time of Electrically-Coupled 
Rooms," /. Acous. Soc. Amer., 4, 1 (July, 1932), p. 63. 



Summary. High-speed motion picture cameras capable of taking pictures on 
continuously moving film at the rate of upward of 3000 frames per sec produce in- 
dividual exposures on the order of V/ooo sec or less. Exposures of such brevity, 
however, call for continuous illumination of great intensity, and incandescent lamps, 
which are adequate, have many disadvantages. Among these are: excessive con- 
sumption of electric power, heavy conductors and connectors, emission of consider- 
able heat, and appliances and reflectors of great bulk and weight. 

Since the actual time during which such high-intensity illumination is required 
seldom exceeds one second, certain flash lamps will provide satisfactory illumination 
when operated in suitably designed equipment. Flashing successively on the current 
of a 6-v dry-cell battery, one or more flash lamps will yield light of high actinic value 
and of easily controlled direction and duration. 

Special equipment, known as the Continuous Flash Lighting Unit, accomplishes 
such purposes effectively, providing adequate illumination both for black-and-white 
and natural color high-speed motion pictures. This paper reviews the development 
of the Continuous Flash Lighting Unit, and describes its operation and advantages. 

Customary Lighting. The problem of illumination grows in 
truly geometric progression with the size of the area to be photo- 
graphed. Each user of the high-speed camera solves this problem 
differently, depending upon the subject under investigation and local 

A most universally successful solution in general is provided by 
the use of "over-volted" incandescent lamps, such as the Mazda R-2 
Reflectorfloods and the Wabash Reflector Superfloods. Both oper- 
ate at high intensities on 115-v current. The GE 150-w projection 
spotlight lamp, although rated at 120 v, provides adequate illumina- 
tion for high-speed photography when operated on 220-v. 

Though relatively short-lived, these lamps provide satisfactory 
illumination for small objects which can withstand the intense heat 

* Presented June 13, 1945, at a meeting of the Atlantic Coast Section of the 
Society in New York. 

** Cinematographer, Editor, Photo-Lab-Index, New York. 



emitted by the lamps when used at close range. These lights must 
be used at distances of 12 to 16 in. from the subject, with their light 
beams superimposed. 

Four Mazda R-2 Reflectorflood lamps placed about 16 in. from the 
subject, with their beams superimposed, will illuminate adequately 
an area approximately 9 in. sq for an exposure of 2500 frames per 
sec, with the lens set to //2.7. The current consumed by these 4 
lamps is 20 amp at 115 v. Voltage should not be allowed to drop 
much below that level, if the exposure is to be acceptable. The 
"bluishness" of the Photoflood lamps, measured by the high color 
temperature at which they operate when over-volted, drops off rapidly 
as the voltage is reduced. A drop of 5 v from the rated level will lose 
as much as 100 K, resulting in considerable loss of the actinic value 
of the lamps. 

Obtaining 20 amp of 115-v current may not be much of a prob- 
lem in most locations. Yet it can be a headache when 4 Reflector- 
flood lamps are inadequate for the illumination of some subjects or, 
for that matter, "too hot to handle" in connection with others. 
The heat emitted by these lamps can have undesirable or adverse 
effects upon some subjects to be photographed. Take the case of the 
center portion of a large aircraft propeller blade which had to be pho- 
tographed in motion (vibrating) : An area some 20 in. sq had to be 
illuminated to a higher than normal level because of certain require- 
ments of the investigation. More than 20 kw of power, and a trans- 
former, had to be used. Incidentally, 20 kw of electric power call for 
No. OR conductor cable. Photographically, the results were en- 
tirely acceptable. Yet the steel blade, housed in a relatively small 
compartment for this investigation, became so hot during the pre- 
liminary focusing, line-up, and several exposures that some doubt 
was cast upon the validity of certain observations made on viewing the 
footage. On that job the cameraman and his associates were nearly 
roasted in more ways than one. 

The Origin of the Idea. A recommended solution to this il- 
lumination problem originated in the mind of a cameraman whose 
assignments vary and whose problems, often hundreds of miles 
apart, have the added distinction of dissimilarity. It suggested itself 
during a high-speed investigation of the action of Photoflash lamps 
synchronized with camera shutters of various types. In this case 
there was no problem of rendering the subject visible. The subject 
itself was self-luminous. 

360 H. M. LESTER Vol 45, No. 5 

The shutters required no illumination because another flash lamp, 
placed behind the shutter, served to illuminate the opening and 
closing of the shutter blades, or the movement of the curtain slit. 
However, owing to the high level of light intensity of these 2 "per- 
forming" flash lamps, their flashes seemed to appear upon the screen as 
coming from nowhere. They bleached all adjacent detail of auxiliary 
equipment, and gave the audience no opportunity to orient itself to 
the relationship of the various parts in the field, or to recognize the 
type, size, and character of the flash lamp shown, or the synchronizer, 
or the shutter. 

It appeared desirable to "pre-illuminate" each performing flash 
lamp in such a manner that it would be seen together with its properly 
arranged accessory equipment for some time before its action started. 
Reflectorfloods were found to be too weak. Since many of them had 
to be used, their multiple reflections upon the glass envelope con- 
cealed too much of the inside of the flash lamp. Backlighting, which 
would have been effective for the illumination of a flash lamp alone, 
could not be used because cameras, shutters, and synchronizers ad- 
joining the flash lamp in the picture had to be lighted too. Ob- 
viously, "the wedge had to be knocked out by means of another 
wedge." A single flash lamp in a reflector was placed several feet 
away from the subject and directed upon the scene. Fired slightly 
before the flash lamps being photographed flashed, it provided the 
necessary illumination, a desirable "fade-in," and a balanced expo- 
sure at//8. 

The preillumination Photoflash lamp was fired by a 6-v current 
closed by the camera's Microswitch, which was pre-set to a pre- 
determined point of the footage. The same contact-maker also 
tripped a delayed action relay, which closed, a little later, the circuit 
of the synchronizer flashing the performing lamps and moving the 
camera shutter. The delay was about 10 milliseconds. Thus, the 
flash lamps and the camera setup, the combined action of which was 
being studied, were first illuminated by the light of a flash lamp, then 
put in action by the simple relay. 

At the high operating speed at which these pictures were taken, the 
light emitted by the flash lamp exposed a considerable footage of film. 
At 3000 frames per sec the camera exposes 3 frames during each milli- 
second. Since this particular flash lamp has a flash duration of some 
60 milliseconds, its light illuminated some 4 l / 2 ft of film. In addi- 
tion, it was noted that the light transition from the preillumination 


lamp to the performing lamps was quite smooth and not apparent to 
the audience. This suggested the possibility of using Photoflash 
lamps, fired in suitable succession, as a source of high-intensity il- 
lumination for high-speed motion picture photography. 

Many Advantages. This possibility appeared immediately as 
most attractive in many respects. Electric current requirements 
are reduced to nothing more than a 6-v dry-cell battery. The heat 
emitted by the flash lamps, being only momentary, is negligible. 
Further, the high blue-light content, or the high color temperature, of 
Photoflash lamps (3800 K) being substantially higher than that of 
Photon 1 ood lamps (3400 K), and being entirely independent of line 
voltage fluctuation since the light output is based upon the uni- 
formity of the inflammable charge promised a light of much higher 
actinic value. 

Other advantages are found. These include smaller lens stops, 
greater depth of field, greater over-all sharpness. Also, it would ap- 
pear that exposure calculations are simplified, heavy cables dispensed 
with, and the entire outfit made quite portable because most of its 
electric power requirements are self-contained. 

The method, however, took some working out. To minimize un- 
evenness and pulsation of illumination and to reduce to a minimum 
the number of flash lamps in a cycle, a lamp was needed which 
produced the longest flash. Such a lamp was found in the Mazda No. 
31 Focal Plane Photoflash lamp, made by the General Electric Com- 
pany and the Westinghouse Company. With peak reaching ap- 
proximately P/2 million lumens, flash duration at one-half peak is 
around 55 to 60 milliseconds (about 1 /2o sec). Since the combined 
brightness of 4 R-2 Reflectorflood lamps adds up to only some 68,000 
lumens, the advantage of the output of a suitable flash lamp cycle is 

Another flash lamp suitable for the same purpose is the Wabash 
No. 2 A Superflash lamp. It has similar performance characteris- 
tics, but slightly different physical dimensions. This lamp cannot 
be substituted for the Mazda lamps, nor should it be mixed with them 
in the same cycle. Either the Superflash or the Photoflash lamps are 
suitable for this purpose. 

The Prototype of the Equipment. Considerable experimentation 
with the No. 31 Photoflash lamps, combined with extensive calcula- 
tions and plotting of the time-light characteristics of the flash lamp 
(Fig. 1) led to the construction of a revolving contactor on an experi- 



Vol 45, No. 5 

mental basis. The unit consisted of a flat panel of aluminum upon 
which were mounted 15 lamp sockets, each wired to a rotating 
"distributor," shown in Fig. 2. A synchronous motor revolving at 
75 rpm rotated a sliding contact, which made one revolution in 
4 /s sec (800 milliseconds). The sliding contact was made to rotate 
continuously and the battery current required for the firing circuit 
was closed at an appropriate moment by a holding relay, activated 
by the Microswitch on the camera footage indicator. 

This prototype of the continuous flash lighting unit was used and 
experimented with quite extensively. Exposures made with it were 
found to have a reasonably satisfactory evenness of illumination. 

10 20 30 


FIG. 1. Focal plane Mazda Photoflash lamp No. 31. 

Views were obtained of larger areas at smaller lens apertures, with 
the lights farther away from the subject than had been possible with 
incandescent light sources. 

However, several drawbacks soon became apparent. The lamps 
were arranged upon the panel in 3 rows of 5 lamps each, spaced rather 
widely, though uniformly. The generous spacing was provided 
partly to gain some reflection of the light from the aluminum panel, 
partly because of the physical dimensions of the lamp sockets, but not 
to avoid flashing of the lamps spontaneously "by contact." Only the 
earlier foil-filled types of flash lamps fired "sympathetically," when in 
contact with other flashing lamps. The wire-filled, or the shredded 
foil-filled lamps, do not act in such a manner. 

Although the lamps were wired in sequence to the distributor, since 


the contactor was turning continuously before the flashing cycle 
started, it was never possible to predict which lamp was going to fire 
first. The gap between the flashes of lamps from opposite corners of 
the rectangle, and the generally broad spacing between the lamps, 
produced a directional shift of the successive light sources. This was 
noticeable in the exposed footage, lights and shadows shifting across 
the subject. The flat reflector proved to be inadequate, for the 
light lacked necessary "punch" and direction. 

FIG. 2. Original experimental continuous flash lighting unit, employing 15 
No. 31 Photoflash lamps. 

Present Equipment. Further experimentation with the flash 
lamps revealed that a much more satisfactory flashing cycle would 
result from the use of 17 flash lamps. These would yield a fairly 
level output of something over l l / 2 million lumens for a full second, 
sufficient time to expose some 60 ft of film at 3000 frames per sec, 
and greater footage at higher speeds of camera operation. 

In the present equipnlent, each of the 17 No. 31 Photoflash lamps 
is held by a special thin-walled slip socket, mounted upon the outer 
circumference of a wheel of some 8 in. in diameter. This forms the 
flashing rotor, which is driven by the slow shaft of a synchronous re- 
ducer motor and revolves at the rate of one revolution per second. 
Its rotation results in the firing of the lamps at intervals of 1 /n sec 
(approximately 59 milliseconds). 



Vol 45, No. 5 

Fig. 3 shows the time-light characteristics of 2 successively flashing 
lamps and the resulting composite curve representing their combined 
light output. The graph shows that there is a time lag of approxi- 
mately 7 milliseconds after the flashing circuit is first closed and be- 
fore the illumination starts. At 10 milliseconds the light level 
reaches about 250,000 lumens; at 14 milliseconds, 500,000 lumens; at 
18 milliseconds, 1,000,000 lumens. Thereafter comes a slower rise 
to I 1 / 2 million lumens and beyond at 59 milliseconds. At this point, 
however, contact is made for the next adjoining lamp, the light out- 
put of which starts at about 66 milliseconds counting from zero 
time just as the light of the first lamp begins to fall off. The 
cumulative effect is shown by the solid line, revealing a momentary 

DEGREES ( 360* to 1000 rmlhseconds ) 
o 5 10 15 20 

25 30 35 40 45 50 


1 '5 






/ \ 





o Oj 

/ ,.: 

x ' 





20 10 20 30 40 50 6 
TIME - milliseconds 

70 80 90 1C 
^ Time- Light C 

110 12 

haractenstic Cu 
Flash Lamp 

130 140 15 


of Smgl 

SOLID line shows Summary Time-Light Performance Characteristic Curve of Flash Lamps firing m suitably spaced succession 
in the Continuous Flash Lighting Unit Total duration for 17 lamps. 1 second 

Fig. 3. Two of the 17 cycles of flashes which occur during one second of 
continuous flashing in the rotary unit. 

dip in the total illumination, followed by a small rise, the composite 
of the rise and fall of the light output of the 2 respective adjoining 
lamps. At 90 milliseconds, the first lamp is completely out, and the 
light level curve is that of a single lamp again. At 110 milliseconds 
the firing contact is made for the third lamp, and the cycle is repeated. 

Obviously, an absolutely even, straight-line level of illumination 
is not attained. However, the variation appears to be within the 
limits of 1.25 to 1.60 million lumens, a difference well within the 
limits of latitude of black-and-white film. Actual experience bears 
out that although discernible, the light fluctuation is not at all objec- 
tionable when the film is viewed. In color films this fluctuation is 
slightly more pronounced, but still quite acceptable. 

The flashing rotor, carrying the flash lamp sockets, contains within 
its underside a large segment commutator serving as the distributor 

Nov.. 1945 



for the firing current. Each lamp socket is connected to its corre- 
sponding bar on the commutator, and contact is made by a brush ex- 
tending from the frame of the motor unit. The order of firing and the 
exact interval between the contacts is thus predetermined and fixed 
by the construction of the commutator. The commutator and the 
contact brush are adjustable with respect to the lamp sockets, so that 
the exact position of the firing point for all 17 lamps can be adjusted 
through an angle of some 15 degrees. This adjustment is useful not 
only for the consummation of the entire flash within the physical 



FIG. 4. 

Top view of continuous flash lighting unit showing relative 
position of lamps on rotor. 

confines of the reflector, but also, when 2 such units are used simul- 
taneously, to level off the resultant light output of both light sources. 
It is possible to set the flashing rotors with respect to each other in 
such a manner that the peaks of one cycle occur during the valleys of 
the other, and vice versa. 

The relation between the flashing lamps and the reflector is shown 
in detail in Figs. 4 and 5. As drawings and photographs indicate, the 
reflector is provided with a cutout gate through which the revolving 
lamps enter and leave the reflector. Ideally, the lamp should make 
contact just as it enters the confines of the reflector, and reach its 
maximum peak of light as it crosses the focal point of the parabolic 



Vol 45, No. 5 

reflector. This adjustment is easily obtained and, once set, is per- 
manently maintained by means of setscrews. 

The motor is started substantially before the camera. The flash- 
ing rotor will revolve before and after the flashing cycle until shut off. 

The reflector is made of a special aluminum sheet, Alzak finished, 
its surface directing upon the subject 5 to 7 times the bare lamp value 
of light. 

The Control and Power Unit. Figs. 6 and 7 show the power and 
control unit, which contains an ordinary 6-v dry-cell (Hot Shot) 

H,gh-Speed Motion Picture Photography ' 

(Patent Applied for) 
by Henry'M Lester Nei York 

FIG. 5. Side view of continuous flash lighting unit showing adjust- 
able support of motor, rotor, and reflector on Y yoke. 

battery. This provides an almost inexhaustible source of electric 
current for the flashing of the lamps. The unit has 2 polarized re- 
ceptacles for 4-conductor Jones connector plugs connecting one or 
two of the continuous flash lighting units through a shielded cable. 
The 4-conductor cable carries the alternating 60-cycle, 115-v current 
to the synchronous reducer motor driving the flashing rotor, and the 
6-v current distributed to the flash lamps through the commutator. 
The control. unit contains, also, the locking relay, which maintains 
the battery current through the commutator until all the 17 lamps 
have been fired after the contact has been made by the Microswitch 


on the camera. The control unit permits the operation of one or 2 of 
the units on a remote control basis. 

Even Illumination. Obviously the evenness of illumination de- 
pends upon all lamps firing uniformly and correctly. Excessive 
resistance in flash lamp circuits results in their delayed firing if, in- 

FIG. 6. Continuous flash lighting unit with parabolic reflector re- 
moved, and control box with ohmmeter used to check flash lamp 

deed, the lamp does not fail to fire. To minimize chances for such 
failures, all wiring is heavier than actually necessary to carry the 
flashing current. Contact brush and commutator bars are quite large 
for the load. However, excessive resistance in any of the 17 lamp 
circuits may be present because of (1) improper seating of a lamp in 
its socket, causing defective contact, (2) corroded or fouled lamp base 
tip, or (3) a defective or damaged lamp. 



Vol 45, No. 5 

Since the high-speed camera used in these tests has only one 
Microswitch for the making or breaking of an external electric circuit, 
the control unit further provides means for firing the lamps and start- 
ing the action of the subject before the camera in any predetermined 
order. This is accomplished by a "delaying" mechanism The con- 
tact made by the camera Microswitch releases a rotating, spring- 
loaded cam. The cam, in turn, 
operates a second Microswitch 
after rotating through an ad- 
justable part of a revolution. 
The number of degrees through 
which the cam rotates before 
making contact is adjustable 
through a range of approximately 
60 milliseconds. This device 
may accordingly be used to (1) 
start the flashing lamps before 
the action to be photographed, 

(2) start the action of the sub- 
ject before the lamps, to allow it 
to reach its operational speed, or 

(3) operate the 2 flashing lamp 
units at varying time intervals 

I instead of having them light up 

High-Speed Cinematography 
with Continuous Flash Lighting 
HHH Units. High-speed camera 
work is considerably simplified 
with continuous flash lighting 
units. Aside from the conveni- 
ence of having the power supply for the unit completely contained 
within the unit, and portable (only sufficient 115-v a. c. is needed 
to operate the small synchronous motor of l / m hp, which can be 
obtained from any household electric outlet), the absence of heat 
and of heavy cables and bulky lightstands greatly simplifies opera- 

Photographically, a 3,000,000-lumen level of light is maintained 
long enough to expose 60 ft or more of film at 3000 frames per sec. 
Two units provide enough light to illuminate amply an average sub- 

FIG. 7. Continuous flash lighting unit 
with "power patch" and control boxes. 


ject for exposure in black-and-white at lens apertures ranging from 
//4 to//8, depending upon conditions. 

Although the flashing units can be safely placed within a few inches 
of the subject, it has been found convenient not to crowd the subject 
by placing them nearer than 3 ft. It is almost always important to 
have free access to the subject before and after it has been photo- 
graphed. The high color temperature of flash lamps (3800 K) 
increases the efficiency of illumination for black-and-white work. 

For use with color films, such as Ansco Color, tungsten type, 
or Kodachrome, type A, continuous flash lighting units afford suf- 
ficient illumination for exposures at 3000 frames per sec at lens 
stops of //2.7 to //4, depending upon conditions. 

It is believed that many users of high-speed cameras currently 
available will find continuous flash lighting illumination of consider- 
able interest. Also, it is possible that as improvements are made in 
this method of illumination, cameras attaining even higher speeds of 
operation will be introduced. It is hoped that broader acceptance 
of the continuous flash lighting method for high-speed camera work, 
resulting in great consumption of flash lamps of this type, will induce 
makers of flash lamps eventually to design them specifically for use in 
continuous flashing. Then the peaks and valleys of the total light 
output can be minimized, and, particularly for color films, this method 
of illumination will provide both the quantity, quality, and evenness 
of light output so desirable. 


EASTMAN KODAK Co.: "An Extensive Bibliography of High-Speed Photog- 

BOON, J. L.: "The Eastman High-Speed Camera, Type III," J. Soc. Mot. 
Pici. Eng., 43, 5 (Nov., 1944), p. 321. 

EDGERTON, H. E., AND GERMESHAUSEN, K. J.: "Stroboscopic-Light High- 
Speed Motion Pictures," J. Soc. Mot. Pict. Eng., XXIII, 5 (Nov., 1934), p. 284. 

EDGERTON, H. E.: "Stroboscopic and Slow-Motion Pictures by Means of 
Intermittent Light," /. Soc. Mot. Pict. Eng., XVII, 3 (Mar., 1932), p. 356. 

LESTER, H. M.: "Photoflash Lamps Motion Pictures," G.E. Review, 47, 
4 (Apr., 1944), p. 19. 

LESTER, H. M.: "A Special Problem in Time Microscopy," /. Phot. Soc. of 
Am., 11, 2 (Feb., 1945), p. 65. 

PALME, A.: "Intense Light," Amer. Phot. (Jan., 1945), p. 17. 

WATSON, E. M.: "Aids for Pictorially Analyzing High-Speed Action," /. 
Soc. Mot. Pict. Eng., 43, 4 (Oct., 1944), p. 267. 

WATSON, E. M.: "Fast Motion Analysis as an Aid to Organized Invention," 
/. Soc. Mot. Pict. Eng., 43, 4 (Oct., 1944), p. 289. 

WESTERN ELECTRIC Co.: "Western Electric Fastax High-Speed Motion Pic- 
ture Camera," Instruction Bulletin No. 1079 (1943). 


Summary. The Western Electric 1100-type densitometer has been modified 
to permit direct recording of H and D curves and other photographic data. A special 
linear 80 db Speedomax recorder permits direct recording of densities from zero to 

At the time of the original introduction of the Western Electric 
RA-1100 densitometer, 1 some consideration was given to the possi- 
bility of providing a direct recording attachment to obviate the 
necessity of plotting characteristic H and D curves obtained from 
density readings of the step tablet sensitometer strip. However, 
since no suitable commercial recording device was available at that 
time, it was decided to introduce the instrument to the industry with 
direct reading facilities, leaving the recording feature to be incorpo- 
rated at a later time whenever suitable equipment might be developed 
for this purpose. The instrument described in this paper was de- 
signed and built to meet the requirements of the Triplett and Barton 
X-ray Testing Laboratories, Burbank, California. 

Anybody familiar with the use of a densitometer in a large produc- 
tion laboratory realizes the prodigious effort put forth in the plotting 
of H and D control strips and will readily agree that automatic plot- 
ting of these curves would not only save a great deal of labor, but 
would at the same time make possible much more accurate graphical 
representation of the data. Since density is a logarithmic function 
of film transmission, the output of the recording device must bear a 
logarithmic relationship to the transmission of the film. There are 
2 possible methods of doing this: either by designing a densitometer 
amplifier to have a logarithmic response, that is, the output to be a 
logarithmic function of the input, or by using a linear amplifier and 
incorporating a logarithmic attenuator to control the movement of 
the recording device. 

* Presented May 15, 1945, at the Technical Conference in Hollywood. 
** Electrical Research Products Division, Western Electric Company, Holly- 


The relationship between decibel ratio and density may be deduced 
as follows : 

Let TI = transmission of one silver deposit 

Let Tt = transmission of another silver deposit 

By definition, the corresponding densities Z?i and Z> 2 are given by 

Di = log l/Ti 
D 2 = log 1 /T t 
or Di - D 2 = log 

The decibel ratio of the light energy transmitted by the 2 deposits is given by 

db = 20 log TV 7\ 
db = 20 (D, - D 2 ) 
For DI = 4.0 and D 2 = 
db = 20 X 4 = 80 

Since the newer model, 1100 A densitometer, is designed to read 
densities from to 4.0, the amplifier output would have to be propor- 
tional to the logarithm of input over an 80-db range. This is very 
difficult, if not impossible, to achieve with present techniques and, 
consequently, was abandoned early in the investigation. Fortu- 
nately, the Leeds and Northrup Company 2 had published an article 
describing a 40-db continuous recording Speedomax, the output as 
indicated by the position of the recording pen, bearing a linear decibel 
relationship to the input voltage to the recorder. Since it would be 
possible to record up to a maximum density of only 2.0 with this de- 
vice, the Leeds and Northrup Company was asked to provide a modi- 
fied Speedomax recorder with a linear 80-db range, thus permitting 
the recording of densities from to 4.0. 

In order to convert the RA-1100 densitometer into a recording de- 
vice, provision must be made for moving the sensitometer strip uni- 
formly past the scanning beam. If the sensitometer strip is of the 
ordinary step tablet type exposed by means of an Eastman lib 
sensitometer, a discontinuous curve would be plotted by the record- 
ing device. If a smooth H and D curve is to be obtained from the 
instrument, provision must of course be made for a continuously 
varying exposure in the sensitometer. 

In the RA-1100 densitometer modified for recording purposes, a 
specially designed high gain amplifier is bridged across the input of 
the standard densitometer amplifier. This permits the use of the 
standard instrument in a normal manner even with the addition of 
the recording attachment. The additional amplifier must be capable 
of delivering a sufficient voltage for operation of the Speedomax 



Vol 45, No. 5 

recorder. Thus, the amplifier output should be such that the re- 
cording pen will assume the zero density position with no film in the 
scanning gate. At the same time, the noise output of the amplifier 
must be so low that the recorder will go off scale on the high-density 
end when the light beam is completely blocked. Aside from a spe- 
cially coupled network between the amplifier and the Speedomax unit 
itself, the modified densitometer requires the addition of the 2 sepa- 
rate items referred to above. These will now be described in some 

FIG. 1. RA-1200 densitometer film moving mechanism. 


A consideration of the problem of pulling film automatically, with 
a device to be applied to existing densitometers, led to the following 
requirements : 

(1} If a continuously variable sensitometer strip is to be used, the scanning 
aperture must have its shortest dimension in the direction of the film travel. 

(2) The film must be driven by a source which is synchronized with the paper 
drive of the Speedomax recorder. 

(5) The film must be controlled in a manner which will provide for rapid and 
convenient operation and also insure accurate density measurements by the 

The optical system of the densitometer is such that the scanning 
beam cannot be readily rotated 90 degrees and, therefore, it seemed 
desirable to move the film toward the operator, which direction is at 
90 degrees to the standard arrangement. This permits a continu- 


ously variable sensitometer strip to be scanned by an aperture which 
is 175 mils wide by either 16 or 65 mils in the direction of the film 
travel. A film drive unit was, therefore, developed to meet the above 
requirements, and arranged to replace the T-shaped film guide 
normally used with the densitometer. This device is shown by Fig. 
1 and installed on the densitometer in Fig. 2. It consists of a film 
pulling assembly mounted on a base plate which is fastened to the den- 
sitometer panel by means of 4 thumbscrews. These may be quickly 
removed to permit the use of the standard film guide. The film drive 
is accomplished by a pair of rollers which propel the film by friction 
and it is designed to operate with film strips 35 mm in width. The 
upper roller is driven by a small synchronous motor with the proper 

FIG. 2. RA-1200 densitometer film moving mechanism and amplifier. 

gearing to drive the film strip exactly 10 in. in 30 sec. A small lever 
on the front of the device is used to lift the upper roller so that the 
film strip may be inserted from the front. With the strip in the 
proper position required for scanning, the lever is lowered to contact 
the film and sufficient spring tension is provided to prevent slippage. 
The synchronous motor is connected to the same source of power as 
the Speedomax paper drive motor and both motors are started by 
the same switch. The accelerating times of the 2 motors are quite 
short and are sufficiently alike that no measurable error is introduced. 
It is essential that good contact be maintained between the film 
emulsion and the scanning aperture plate of the densitometer. This 
is accomplished by 2 small spring-mounted rollers which bear on the 
film on both sides of the aperture. To protect the film from dirt or 
abrasion, a protective guide assembly is inserted in the slot under the 



Vol 45, No. 5 

densitometer head to receive the film as it is pushed in from the front. 
It is lined with plush and so constructed that it may be opened for 

For fast and convenient operation with this device, it is recom- 
mended that the continuously variable sensitometer strip conform to 
the requirements shown on Fig. 3. This proposed sensitometer 
scale is 10 in. long and starts at a punched hole which serves as an 
index which will register with the scanning beam as the operator 
places the strip in the instrument. This enables the operator to 
watch the meter of the densitometer which gives a zero or minimum 
reading when the punch mark is in exact register with the scanning 




FIG. 3. Sensitometer strip specifications for use as RA-1200 recording 

aperture. When this condition is attained, he drops the lever which 
engages the film with the driving roller and the device is then ready 
for use. 

The film pulling assembly may be shifted laterally on its base plate 
by turning a knob so that the film may be scanned at any point to 
within 3 /i 6 in. of either edge. The scanning position is indicated by 
an index mark and scale as shown by the photograph. It is assumed 
that the X-ray sensitometer strips would normally be scanned at the 
center, but this facility is provided to accommodate other types of 
sensitometer strips such as those encountered in motion picture prac- 

The densitometer must be slightly modified to accept this device, 
the modification consisting primarily of cutting an opening in the 



panel to clear the lower roller assembly of the film pulling mechanism. 
Four holes are tapped in the panel to accept the mounting thumb- 
screws and the slot in the head casting is widened to permit the lateral 
shift of the film. The T-shaped film guide is built out so that after 
modification, the densitometer may be used either with this device 
or in the normal manner, and it may be assembled or disassembled 
in 2 or 3 min. 


Since it is desirable to avoid the necessity of adding an extra power 
supply for the added amplifier, it was necessary to design the latter so 


FIG. 4. Extra amplifier required to operate Speedomax recorder. 

tha.t it would operate satisfactorily from the power supply originally 
furnished with the densitometer. Since this power supply furnishes 
180 v plate supply, vacuum tubes had to be selected and an output 
circuit designed to operate from it to furnish the necessary power for 
operating the recording unit. The amplifier is connected through a 
plug connection to a pair of input terminals of the densitometer am- 
plifier, shown in Fig. 4. The first stage is a phase inverter for coup- 
ling to the output push-pull circuit, which utilizes a pair of 6 V6 vac- 
uum tubes. The gain of the amplifier is controlled by the potenti- 
ometer in the input circuit. Pressing the calibration button auto- 
matically inserts an 80-db loss into the amplifier for checking the 4.0 
density point on the Speedomax scale. In addition to furnishing 



Vol 45, No. 5 

the necessary power to operate the Speedomax recorder, the amplifier 
must have a linear relationship between output and input over the 
80-db range, and the signal-to-noise ratio must be of the order of 90 
db. Both of these requirements have been substantially met. 

The output of the amplifier described above is coupled through a 
high-pass filter and isolating transformer to the input of the Speedo- 
max unit. The purpose of the high-pass filter is to attenuate low- 
frequency microphonic noises and a-c hum which would result in 
spurious readings. Since it is only necessary to transmit the 400-cycle 
frequency generated by the densitometer chopper wheel, the addition 

FIG. 5. Complete RA-1200 recording densitometer. 

of a low-pass filter cutting off sharply above this frequency would 
further enhance the signal-to-noise ratio of the system. However, 
the operation appears to be quite satisfactory without the addition 
of such a filter. The high-pass filter and isolating transformer are 
mounted on the framework of the Speedomax unit. 


The operation of the Speedomax recorder has been described else- 
where 2 so that it will not be necessary to go into any further detail 
here. For the benefit of those not familiar with the operation of this 
unit, it is only necessary to state that the device operates on the self- 


balancing principle. To illustrate, assume that there is no film in 
the scanning gate and that the recorder indicates zero density. The 
insertion of any density in the gate reduces the input to the Speedo- 
max unit. This disturbs the balance within the unit and causes a 
motor to drive the input logarithmic potentiometer to a new position 
at which the balance is again restored. The same motor which drives 
the attenuator also drives the recording pen, and since the movement 

FIG. 6. Speedomax recorder associated with 
RA-1200 recording densitometer. 

of the attenuator arm is linear in decibels, the movement of the pen 
will also be linear in decibels or density. If a density of 4.0 is in- 
serted in the scanning gate, the pen will move across the paper from 
the position of zero density to that of 4.0. For any intermediate 
value of density, the pen will stop at the appropriate position on the 
scale. The recorder has sufficient speed and damping that it will 
record an abrupt full-scale change in less than 2 sec without over- 



Vol 45, No. 5 

The paper chart on which the photographic characteristics are re- 
corded is driven by a synchronous motor insuring that the paper 
moves at the same speed as the film being drawn past the scanning 
aperture. Since the log exposure axis is parallel to the direction of 
the movement of the paper, the scale will be determined by the linear 
speed of the latter. While it is desirable that the density scale across 
the paper be identical with the log E scale, in order to permit direct 
reading of contrast or gamma, it is not essential since the value of 

i i ii i 1 1 1 1 1 1 1 1 1 1 ii 1 1 i 1 1 1 1 1 1 i 1 1 1 1 1 1 1 1 


FIG. 7. Relationship between density and exposure on a 
radiographic film. 

gamma may be obtained from a suitably engraved template. In the 
present instrument the log E scale extends over a distance of 10 in. 

Fig. 5 shows a photograph of the complete densitometer with re- 
cording attachment, while Fig. 6 is a photograph of the Speedomax 
recorder alone. Fig. 7 is a chart made by Triplett and Barton show- 
ing the relationship between density and exposure on a radiographic 
film. In this case the exposure on the film is made through a tri- 
angular shaped wedge of metal so that a continuously varying log 
exposure scale is achieved. The complete chart was made in 30 sec. 

The recording sensitometer described in this paper has been in 


service for several months and has proved to be very reliable in opera- 
tion, as well as rendering a true graphical representation of the photo- 
graphic characteristics under investigation. Its extended use in the 
future will largely eliminate the time-consuming plotting of H and D 
curves and other photographic characteristic data, thus adding to 
the production value of the RA-1100 type densitometer. 


1 FRAYNE, J. G., AND CRANE, G. R.: "A Precision Integrating Sphere Den- 
sitometer," /. Soc. Mot. Pict. Eng., XXXV, 2 (Aug., 1940), p. 184. 

2 CLARK, W. R.: "The 'Speedomax' Power-Level Recorder," Trans. A. I. 
E. E., 59 (Dec., 1940), p. 957. 



Summary. The variable-density and variable-area systems of recording sound 
on film, with the many available forms of standard or push-pull tracks, enable the 
sound engineer to select the type of sound track best suited to his particular problems. 

A combination of 200-mil variable-density push-pull tracks for original recording 
and 100-mil variable-area standard tracks for theater release has been successfully 
employed by the Sound Department of Columbia Studios on a number of productions. 
This combination has been found to provide high output in the theater with low over- 
all distortion values. 

The processing of standard variable-area tracks in different release laboratories 
has long been a critical problem. A method is described which greatly simplifies the 
solution of this problem and which, in its application, has proved to be thoroughly 
practical under widely varying conditions. 

The 2 fundamental systems of recording sound on film, namely, 
the variable-density and the variable-area systems, with the many 
available forms of standard or push-pull tracks, enable the recording 
engineer to select the type best suited to the conditions under which 
it is required to operate. Such a statement, of course, ignores the 
practical limitations which may be imposed by recording license 

A combination of 2 such systems, the 200-mil variable-density 
push-pull system for studio recording, and the 100-mil standard 
variable-area system for theater release, has been used intermit- 
tently by the Sound Department of Columbia Studios for some time. 
This combination has provided us with greater latitude for the re- 
lease of our sound than was available to us previously, and it appears 
to meet our recording requirements quite adequately. Since we 
contemplate standardizing on this combination eventually, the fac- 
tors which have prompted this decision may be of interest. They are 
as follows : 

The maximum signal-to-noise ratio of a system is determined by 2 

* Presented May 15, 1945, at the Technical Conference in Hollywood. 
** Columbia Pictures Corporation, Hollywood. 



factors the ceiling, or highest practical limit of modulation, and 
the threshold, or lowest practical limit of modulation. The ceiling 
is fixed by the amount of distortion that can be tolerated at high 
modulations, and the threshold is determined by the highest per- 
missible noise level. 

Considering the 200-mil push-pull type of variable-density record- 
ing, we note that the 200-mil width, plus the application of normal 
noise reduction, pre- and post-equalization, and the use of fine-grain 
film stocks, has increased the signal-to-noise ratio of this system to a 
figure well over 50 db which, in our opinion, is quite adequate for 
studio recording purposes. Reduction of second harmonic distortion 
through action of the push-pull system and the use of limiters to pre- 
vent the highest modulations from intruding excessively into the 
nonlinear portion of the film characteristic have resulted in a total 
harmonic distortion content for our operations that is not greater 
than l x /2 per cent, or an equivalent intermodulation value of 6 per 
cent. These values are based upon a 400-cycle signal modulated 80 
per cent. 

In choosing this system for studio recording we were, to a con- 
siderable extent, guided by our own ideas as to the desirable char- 
acteristics of a recording system. These ideas are based on the 
philosophy that it should be possible to faithfully register the original 
signal upon the recording medium with as little alteration as possible, 
preserving not only the essential elements of the normal frequency 
spectrum of the signal, but also its original volume range. The 
Western Electric 200-mil variable-density system has been found 
well suited to this purpose. 

Another consideration which caused us to favor variable-density 
for studio recording is that we feel this type is somewhat safer from 
the film processing standpoint. In the event of unavoidable ir- 
regularities in the processing operations, the low modulations of the 
variable-density system do not suffer to the same extent as they do in 
the variable-area system, and in addition the density tracks are less 
affected by the dirt and scratch noise of the film. 1 

Certain differences exist between the requirements of original 
recording and those which govern the release of the sound for use in 
the theater. This necessitates a considerable alteration of the original 
product which can be suitably carried out during the rerecording 
operations. One of the chief requirements of the release system is 
that it must have adequate output so that the highest passages of 

382 J. P. LlVADARY AND S. J. TWINING Vol 45, No. 5 

music can be successfully reproduced without adding materially to the 
distortion that already exists in the original material. Release 
prints should also be capable of playing close to the average fader 
setting of the theater toward which most projectionists normally 

We find that the 100-mil standard variable-area system offers an 
excellent medium for this purpose. If we compare the output of a 
standard area track with that of an equivalent density track, we find 
that the output of the area track is 7 db higher, which places it on a 
par with the output of the 200-mil push-pull density track used for 
studio recording. It is a recognized fact that, for the same film out- 
put, the variable-area track has less distortion than the corresponding 
100-mil variable-density track which must be overloaded by 7 db in 
order to obtain an equivalent output. 

Our experience with the 100-mil variable-area sound track has been 
limited to results obtainable from the Western Electric 200-mil 
variable-area modulator described by Benfer and Lorance. 2 We 
have used half of the 200-mil sound track obtained from this Class A 
push-pull modulator as a standard release track. Tests with this 
modulator, using an input of 400 cycles at 80 per cent modulation, 
have shown that the distortion of the original 200-mil push-pull den- 
sity recording is not increased by more than one per cent total har- 
monics in rerecording to 100-mil area release track. This places the 
over-all distortion of the release product at not more than 2 x /2 per 
cent total harmonics when measured with the General Radio Har- 
monic Analyser. 

The variable-area system is excellently adapted to handling reason- 
able excursions into the overload region when rerecording the sound 
effects used in release. Here, the increased amount of distortion 
resulting from the overload is not objectionable and the satisfac- 
tion resulting from the increased output is very gratifying. 

Mueller 3 and Fletcher 1 have shown that the volume range of the 
release track must be reduced on account of the higher threshold of 
noise encountered in theaters. This correction is conveniently sup- 
plied in the variable-area system by means of the electronic com- 
pressor. We find that a fixed amount of compression of the order of 3 
to 5 db into 25, in addition to the normal manual compression re- 
sulting from the habits of our rerecording mixers, is adequate for the 
transfer of the variable-density originals to variable-area release 


It is interesting to note that, in our procedure, the small amount 
of fixed compression supplied by the electronic compressor is intro- 
duced past the point of monitoring in the rerecording channel and is 
consequently inaudible to the rerecording mixer. Later, when the 
mixer hears his work reproduced in the theater, he is satisfied with 
the values of the return. We have interpreted this as an indication 
that the noise level of our dubbing stage is lower than that encoun- 
tered in the average theater and that this method of introducing the 
required amount of fixed compression is thoroughly satisfactory 
once the basic noise level relationships between the dubbing stage 
and theater have been established by practical observation. 

The advantages possessed by the variable-area track when used 
for release must be weighed against certain disadvantages. For 
instance, extremely low modulations may, at times, result in "blasti- 
ness" if the modulations happen to ride on a poorly illuminated por- 
tion of the scanning slit. This condition may be remedied to some 
degree by masking off one-half of the bilateral track and simultane- 
ously increasing the level of the signal by 6 db for such portions of 
the rerecording that require very low level output. This technique 
was originally employed in the Sound Department of Columbia 
Studios and has been patented. 

Our experience in the use of the variable-area system as the medium 
of release has brought to our attention the fact that the means of 
determining optimal processing conditions, through the use of the 
modulated high-frequency method as described by Baker and Robin- 
son, 4 does not provide an adequate solution for the problems arising 
out of the great variety of developing and printing conditions en- 
countered in the various release laboratories. At times, particularly 
in connection with the rerecording of sharp sibilants, we have had 
great difficulty in establishing processing conditions for the area 
tracks that would resolve these sibilants satisfactorily despite the 
fact that we were able to show cancellations of the order of 45 to 
55 db by the modulated high-frequency method. 

In an effort to surmount this difficulty, we have established a 
method of determining optimal processing conditions which we have 
designated as a Sibilant Pattern Test. Although this method has 
not as yet been theoretically analyzed so that it might be presented 
in formal manner, it has proved so practical in solving our processing 
problems that, pending its theoretical investigation which is at 
present under way, we would like to present it in its practical form as 

384 J. P. LlVADARY AND S. J. TWINING Vol 45, No. 5 

it is now used. The test is carried out quite simply and involves no 
greater effort than is required to make the modulated high-frequency 
test. It has the advantage, however, that no special equipment is 
required and the determinations can be made wherever standard 
projection equipment is available. The test may be carried out as 

A short piece of original recording is selected which contains several 
moderately overloaded sibilants abundant in frequencies at the 
extreme upper range of the reproduced spectrum. A print containing 
about 10 ft of this material is made into a loop with an interval of 
unmodulated track between the ends. The loop is then placed in 
the projection machine and the test dialogue is rerecorded continu- 
ously at a series of lamp currents. separated by intervals of 1 /i amp. 
A continuous negative is thus created containing a series of record- 
ings of the dialogue material in the loop, each one of which will be at a 
different negative density. 

In order to obtain the required information on negative densi- 
ties, the loop is made up with sufficient unmodulated spacing be- 
tween the ends of the dialogue to permit the recording of unbiased 
sections suitable for making density measurements. The negative is 
then developed at the standard speed and gamma which has been 
established for this purpose and which will be maintained for all sub- 
sequent development runs governed by this test. After completion 
of this operation, the negative is then used to make a series of prints 
of about 5 different densities spread out over the useful printing range. 
Upon completion of the prints they are identified and the density of 
each is measured and recorded. 

It is obvious from the above that a sibilant pattern test can be set 
up which, with suitably spaced intervals, will cover all practical com- 
binations of negative and print densities within the limits of the 
recording and printing equipment. 

A form similar to that shown in Fig. 1 may now be marked off on a 
sheet of paper with the measured negative densities listed in a vertical 
column and the print densities listed horizontally. A space is pro- 
vided in the included form in which to record the goodness of each 
combination of negative and print densities. The prints are as- 
sembled in the order of running to correspond with the form sheet and 
then projected in the review room. As each section is reproduced 
it is judged for its ability to resolve the sibilants of the test material, 
and each section is given a relative rating using some such convenient 

Nov., 1945 



symbols as VG for very good, G for good, F for fair, B for bad, and 
VB for very bad. 

The tabulated results of the test may then be transferred to a 
graph, shown in Fig. 2, on which the ordinates are print densities 
and the abscissae are negative densities. The symbols VG, G, F, 
etc., are then marked on the graph sheet at the positions correspond- 
ing to their associated negative and print densities and a straight line 
is drawn so as to satisfy the positions of highest rating, such as the 
points of VG if such are used, but also with due regard to satisfying 
the conditions of falling approximately in the mid-position of the F 
points, for example, if this symbol has been used to denote the lower 
limit of what might be considered an acceptable treatment of the 










































FIG. 1. Form for Sibilant Pattern Test. 

sibilant. It is obvious that the lines denoting the outer limits of 
acceptable treatment also are indicative of the tolerence inherent in 
the system of processing under examination, and thus serve a very 
useful purpose for comparison with other systems that may be ex- 

It may seem at first glance that the method employed in the test, 
of listening and grading the results, would not offer the precise deter- 
minations that are obtained by a system of instrumental measure- 
ments such as would be the case in the use of the modulated high-fre- 
quency method of testing. While this may be true in comparing the 
results of tests made under widely separated intervals with unre- 
lated material, the sibilant pattern test nevertheless has proved to be 
extremely practical, not only in the ease with which a competent 
technician can rate the sections and complete the pattern sheet, but 


J. P. LlVADARY AND S. J. TWINING Vol 45, No. 5 

also in the thorough practicability of the results as applied to the 
actual production work carried out under the specifications thus es- 

As in the case of the modulated high-frequency test method, the 
data of the sibilant pattern test so far tabulated do not permit the 
selection of any optimum combination of negative and print densities 
from the infinite number of favorable combinations contained within 
the limits of the pattern. Such optimum values may be determined, 
however, by further editing the prints so that they contain only those 
sections which have previously been given the arbitrary auditory 








-46 CB. 

9000 /4< 

^0 CYCL 










* d 

"> * 








X * 





26 oa 









} (j 











2.7 2.8 2.9 3.0 3.1 





FIG. 2. Optimal processing curves for variable-area recording negatives on 
Eastman 1371 stock. Prints on Eastman 1302 stock. Negatives and prints 
exposed with unscreened tungsten. 

ratings of G and VG, or whatever ratings have been applied to the 
combinations of best resolution under the circumstances of the test. 

These edited tests are then assembled, reviewed, and given com- 
parative ratings, a task which is greatly simplified by the elimination 
of the inferior material. Optimum values of negative and print den- 
sities may thus be determined. If the highest rated combinations 
under this examination do not show a sufficient difference so that 
they may be rated among themselves, it may be concluded that the 
optimum position is noncritical and suitable conditions of matched 
negative and print densities may be selected based on practical con- 
siderations affecting the operation of the printers and recorders. 

When it is desired to compare the results of the sibilant pattern 
test with those of the modulated high-frequency test, both tests 


may be made consecutively on the same piece of negative stock. 

Such comparative tests have been carried out. The results of 
these tests have shown that, in the several cases which were ex- 
amined, the slope of the graph relating suitable negative and print 
densities as determined by each method of testing was the same. In 
certain ideal cases, in which excellent developing and printing condi- 
tions with ultraviolet exposures were available, the graphs of the 2 
tests coincided exactly. Under other circumstances, presumably 
where less suitable conditions of processing obtained, the lines for the 
optimal combinations on the graph sheet were widely displaced. 
These separations have been found to be as high as 0.40 in print 
density in some cases. Under such circumstances, when the speci- 
fications for the film processing were set up on the basis of the modu- 
lated high-frequency tests, the practical application of these values 
resulted in failure to resolve the sibilants adequately. On the other 
hand, when the specifications were based on the results of the sibilant 
pattern tests, the practical operations resulting from the use of the 
values thus determined were found to be entirely successful. 

It has also been found useful to rerecord the sibilant test loop at 
the end of each reel intended for release. When printed, the quality 
of this end test immediately certifies the condition of the reel as a 
whole. To obtain final and precise print densities suitable for re- 
lease printing, the sibilant tests may be removed from the end of 
each release negative, properly identified, and spliced together into a 
single roll. The roll of negative can then be printed to 4 or 5 differ- 
ent densities. The prints are projected and each section given a 
quality rating. Coordination of these data permits a precise deter- 
mination to be made for the most suitable print density for each reel. 
This end test is also useful in the case where a single reel may be 
designated for release in a laboratory where the optimal processing 
conditions have not been determined by a sibilant pattern test. In 
such a case, the end test can be used and a determination made for the 
best print density with the minimum amount of time and expense 
and without the use of special equipment. 

Actually, the sibilant pattern test, which from its nature bears 
directly upon the practical solution of the sibilant problem in vari- 
able-area processing, has proved to be thoroughly reliable and we now 
use it in preference to the modulated high-frequency method in the 
determination of optimal processing values for negatives and prints. 



1 FLETCHER, H.: "The Stereophonic Sound-Film System General Theory," 
/. Soc. Mot. Pict. Eng., XXXVII, 4 (Oct., 1941), p. 331. 

2 BENFER, R. W., AND LORANCE, G. T.: "A 200-Mil Variable- Area Modula- 
tor," /. Soc. Mot. Pict. Eng., XXXVI, 4 (Apr., 1941), p. 331. 

3 MUELLER, W. A.: "Audience Noise as a Limitation to the Permissible 
Volume Range of Dialogue in Sound Pictures," /. Soc. Mot. Pict. Eng., XXXV, 
1 (July, 1940), p. 48. 

4 BAKER, J. O., AND ROBINSON, D. H. : "Modulated High-Frequency Record- 
ing as a Means of Determining Conditions of Optimal Processing," /. Soc. Mot. 
Pict. Eng., XXX, 1 (Jan., 1938), p. 3. 



The 1945 Journal Award, given annually for the most outstanding paper 
originally published in the JOURNAL during the preceding year, was awarded to 
C. J. Kunz, H. E. (Goldberg, and C. E. Ives for their paper "Improvement in 
Illumination Efficiency of Motion Picture Printers." President D. E. Hyndman 
presented the recipients with inscribed parchment certificates at the Dinner- 
Dance held on October 16 during the 58th Semi-Annual Technical Conference 
at the Hotel Pennsylvania, New York. 

The paper was published in the May 1944 JOURNAL and was first presented 
before the Society by the authors, all of Kodak Research Laboratories, Rochester, 
at the October 1942 Technical Conference in New York. A biographical sketch 
of the Journal Award winners is being prepared and will be published in an 
early issue of the JOURNAL. 


Announcement was made during the Technical Conference of the election of 
10 Active Members of the Society to the grade of Fellow by action of the Board 
of Governors in recognition of their contributions to the advancement of the 
motion picture industry and of their services to the Society. Appropriate cer- 
tificates were presented by President Hyndman to seven who were present at 
the Dinner-Dance on October 16. Those elevated are: 

Lieut. Col. John O. Aalberg, formerly with RKO Studios, Hollywood. 

Herbert Barnett, International Projector Corporation, New York. 

John G. Bradley, Director, Motion Picture Project, The Library of Congress, 


George L. Carrington, Altec Service Corporation, New York. 
Major Lloyd T. Goldsmith, Director, Pictorial Engineering and Research 

Laboratory, Signal Corps Photographic Center, Long Island City. 
Charles F. Horstman, RKO Service Corporation, New York. 
Charles E. Ives, Eastman Kodak Company, Rochester. 
C. L. Lootens, Paramount Pictures, Inc., Hollywood. 
G. T. Lorance, International Projector Corporation, New York. 
Jack A. Norling, Loucks and Norling Studios, New York. 


Proposed amendments of By-Laws I, VII, and XI of the Constitution and 
By-Laws of the Society, as published on page 246 of the September JOURNAL, 
were discussed and voted on by qualified members present at a general business 
session of the Society on October 15 during the recent Technical Conference at 
the Hotel Pennsylvania, New York. All were approved and are now in effect. 



As a result of the recent elections, the following is a list of the Officers and 
Governors of the Society for terms beginning January 1, 1946: 

* President: DONALD E. HYNDMAN 

* Past-President: HERBERT GRIFFIN 

* Executive Vice-President: LOREN L. RYDER 
** Engineering Vice-President: J. A. MAURER 

* Editorial Vice-President: ARTHUR C. DOWNES 
** Financial Vice-President: M. RICHARD BOYER 

* Convention V ice-President: WILLIAM C. KUNZMANN 

* Secretary: CLYDE R. KEITH 

* Treasurer: EARL I. SPONABLE 
Governors from the Atlantic Coast Area: 



Governors from the Pacific Coast Area: 



Officers and Managers of the Atlantic Coast Section for terms beginning 
January 1, 1946, are: 

* Chairman: FRANK E. CAHILL, JR. 

* Past-Chairman: CLYDE R. KEITH 

* Secretary-Treasurer: JAMES FRANK, JR. 




Officers and Managers of the Pacific Coast Section for terms beginning Janu- 
ary 1, 1946, are: 

* Chairman: HOLLIS W. MOYSE 

* Past- Chair man: CHARLES W. HANDLEY 

* Secretary-Treasurer: SIDNEY P. SOLOW 

Managers: ** G. M. BEST * F. L. EICH 



One of the largest attended meetings of the Pacific Coast Section of the Society 
was held on September 25 when Major Roy Seawright and Warrant Officer 

* Term expires December 31, 1946. 
** Term expires December 31, 1947. 


Jack R. Glass discussed the preparation of briefing films used by B-29 bomber 
crews. These films, which were made by the First Motion Picture Unit at Culver 
City under the direction of Major Seawright, and which played a vital role in 
pin-point bombing, showed the crews how their course to and over Japan appeared 
from the plane, through the bombsight, and on the radar screen. 

The briefing films had to be made quickly and accurately without resorting 
to reconnaissance flights over Japan which might endanger the missions by re- 
vealing intended routes. Warrant Officer Glass described the project in detail 
and told of creating a large reproduction of Japan which was photographed to 
simulate the appearance of Japanese territory under flight. Samples of the films 
were screened. 

As the second feature of the meeting, the German musical color motion picture, 
Lady of My Dreams, was exhibited. This picture, produced in 1942 using Agfa- 
color, was of interest to Section members and guests because it demonstrated the 
use of an integral tripack color system for 35-mm release prints with sound. 
The print was obtained through the courtesy of the Army Pictorial Service and 
the Academy of Motion Picture Arts and Sciences. 

The meeting was held on the scoring stage of RCA Victor in Hollywood. 


The 58th Semi-Annual Technical Conference held in New York on October 
15-17 was the most successful meeting ever conducted by the Society. Over 300 
registrations were recorded for the 3 -day Conference which required 7 technical 
sessions to present all the papers submitted to the Papers Committee. Individual 
sessions were well attended and interest in all papers presented was unusually 

Because of the decision of the Board of Governors to hold a regular national 
meeting following V-J Day, instead of a local metropolitan area meeting as 
planned previously, there were necessarily some changes made in the scheduling 
of papers as listed in the final printed program. Therefore, in answer to many 
requests which have been received for copies of the program, the editors are 
publishing in this issue of the JOURNAL the complete program as followed for the 
3-day Conference. All technical sessions were held in the Salle Moderne of the 
Hotel Pennsylvania, New York. 

Monday, October 15, 1945 

2: 00 p.m. Opening Session of Conference. . 

Clyde R. Keith, Chairman 

Session opened with a 10-min pre-release 35-mm motion picture 

Welcome by President Donald E. Hyndman. 


Report of the Convention Vice-President, W. C. Kunzmann. 

"The Waller Flexible Gunnery Trainer," by Fred Waller, Vitarama 
Corp., Long Island City, N. Y. 

"A Wide Angle 35-Mm High-Speed Motion Picture Camera," by 
J. H. Waddell, Bell Telephone Laboratories, New York. 

"Nonintermittent Motion Picture Projector with Variable Magni- 
fication," by F. G. Back, Research and Development Laboratory, 
New York. 

"A New 16-Mm Buzz Track Recorder," by M. G. Townsley, Bell 
and Howell Co., Chicago, 111. 

"Use of the 16-Mm Motion Picture in the Educational Recondition- 
ing of Hospitalized Combat Veterans," by E. W. Schultz, Army 
Medical Center (Walter Reed General Hospital), Washington, 
D. C. 

"A Film Splicing and Repair Machine," by Armour Wallingsford, 
Republic Studio, North Hollywood, Calif. 

"A Laboratory Film Ink," by Armour Wallingsford, Republic 
Studio, North Hollywood, Calif. 

"A National Film Library The Problem of Selection," by J. G. 
Bradley, Director, Motion Picture Project, The Library of 
Congress, Washington, D. C. 

8:00 p.m. Evening Session. 

Nathan D. Golden, Chairman 

Session opened with a 10-min pre-release 35-mm motion picture 

Talk on I. G. Farben Afga Color by Lt. Col. R. H. Ranger, U. S. 

Army Signal Corps, on behalf of the U. S. Technical Industrial 

Intelligence Committee. Program was sponsored by the U. S. 

Department of Commerce. 

Tuesday, October 16, 1945 
10:00 a.m. Morning Session. 

Earl I. Sponable, Chairman 

Session opened with a 10-min pre-release 35-mm motion picture 

"The Use of Desiccants with Undeveloped Photographic Film," by 
C. E. Ives and C. J. Kunz, Eastman Kodak Co., Rochester, N. Y. 

"Measurement and Control of Dirt in Motion Picture Processing 
Laboratories," by-N. L. Simmons, Eastman Kodak Co., Holly- 
wood, and A. C. Robertson, Eastman Kodak Co., Rochester, N. Y. 

"Aluminum and Chromium as Gelatin Hardeners," by H. L. Baum- 
back and H. E. Gausman, Paramount Pictures, Inc., Hollywood. 

"An Application of Direct-Positive Sound Track in 16-Mm Release 
Processing by Duplication Method," by Major G. C. Misener 


and G. Lewin, Signal Corps Photographic Center, Long Island 
City, N. Y. 

"A Simplified All-Purpose Film Recording Machine," by G. R. 
Crane and H. A. Manley, Electrical Research Products Division, 
Western Electric Company, Hollywood. 

2: 00 p.m. Afternoon Session. 

Herbert Griffin, Chairman 

Session opened with a 10-min pre-release 35-mm motion picture 

"The Wartime Record and Post- War Future of Projection and 

Sound Equipment," by A. G. Smith, National Theatre Supply, 

New York. 
"Carbon Arcs for Motion Picture and Television Studio Lighting," 

by F. T. Bowditch, M. R. Null and R. J. Zavesky, National 

Carbon Co., Cleveland, Ohio. 
"Westrex Standard Sound Film Reproducer," by G. S. Appelgate, 

Western Electric Export Corporation, New York, and J. C. 

Davidson, Electrical Research Products Division, Western Elec- 
tric Company, Hollywood. 
"Westrex Master Sound Film Reproducer," by G. S. Appelgate, 

Western Electric Export Corporation, New York, and J. C. 

Davidson, Electrical Research Products Division, Western Elec- 
tric Company, Hollywood. 
"Projection Equipment for Review Rooms," by H. J. Benham, 

Radio Corporation of America, RCA Victor Division, Camden, 

"Specialized Photography Applied to Engineering in the Army Air 

Forces," by Major P. M. Thomas and Capt. C. H. Coles, Air 

Technical Service Command, Wright Field, Ohio. 
"A Loudspeaker for Critical Monitoring," by Major P. W. Klipsch, 

Southwestern Proving Ground, Hope, Ark. 
"An Absolute Method for the Determination of the Effective/ Stop 

Calibration of Camera Lenses," by C. R. Daily, Paramount 

Pictures, Inc., Hollywood. 
"A Test Reel for Television Broadcasting Stations," by M. R. 

Boyer, Photo Products Div., E. I. du Pont de Nemours & Co., 

Parlin, N. J. 

8:30 p.m. Georgian Room: Fifty-Eighth Semi- Annual Dinner-Dance. 
President Donald E. Hyndman, presiding. 

Introduction of Officers-elect for 1946. 
Presentation of certificates to Fellows-elect. 
Presentation of Journal Award for 1945. 
Music an4 entertainment until 1:30 a.,m, 



Vol 45, No. 5 

Wednesday, October 17, 1945 
10:00 a.m. Morning Session. 

Major Lloyd T. Goldsmith, Chairman 

Session opened with a 10-min pre-release 35-mm motion picture 

"The Dual Projector," by T. C. Hoad, General Theater Supply Co., 

Toronto, Canada. 
"A Simplified Recording Transmission System," by F. L. Hopper 

and R. C. Moody, Electrical Research Products Division, Western 

Electric Co., Los Angeles, Calif. 
"A Brief for Film in Television," by John Flory, of Grant, Flory 

and Williams, New York. 
"A Complete Motion Picture Production Plant for Metropolitan 

New York," by R. B. Austrian, RKO Television Corp., New York. 
"The Illusion of Depth in Motion Pictures," by Lt. H. T. Souther, 

U. S. Army Signal Corps, Long Island City, N. Y. 
"Theory and Practice of Lighting for the Camera," by Lt. H. T. 

Souther, U. S. Army Signal Corps, Long Island City, N. Y. 

2:00 p.m. Afternoon Session. 

E. Allan Williford, Chairman 

Session opened with a 10-min pre-release 35-mm motion picture 

"Report of the SMPE Color Committee," by J. A. Ball, Chairman, 

"The Printing of 16-Mm Kodachrome Duplicates," by R. M. Evans, 

Eastman Kodak Co., Rochester, N. Y. 
"An Improved Film Drive Filter Mechanism," by C. C. Davis, 

Electrical Research Products Div., Western Electric Company, 

"The Ansco Colorpak Process for Professional Motion Pictures," 

by H. H. Duerr and H. C. Harsh, Ansco, Binghamton, N. Y. 
"Phototube for Dye Image Sound Track," by A. M. Glover and 

A. R. Moore, Radio Corporation of America, Lancaster, Pa. 
"Preliminary Sound Recording Tests with Variable-Area Dye 

Tracks," by R. O. Drew and S. W. Johnson, Radio Corporation 

of America, RCA Victor Division, Indianapolis, Ind. 
"The Behavior of a Blue-Sensitive Phototube in Theater Sound 

Equipment," by J. D. Phyfe, Radio Corporation of America, 

RCA Victor Division, Indianapolis, Ind. 

8:00 p.m. Evening Session. 

Glenn E. Matthews, Chairman 

Session opened with a 10-min pre-release 35-mm motion picture 


"The Problem of Amateur Color Photography," by R. M. Evans, 
Research Laboratories, Eastman Kodak Co., Rochester, N. Y. 

Note: This was a one hour and forty-five minute semi-popular lecture 
and demonstration on color photography. 

Adjournment of Fifty-Eighth Semi-Annual Technical Conference, 
by D. E. Hyndman, President. 



Designer and engineer experienced in optics, lighting, and microphotog- 
raphy, capable of designing microfilm reading equipment and products 
related to microfilm industry. Reply to Microstat Corporation, 18 
West 48th St., New York 19, N.Y. 

Design engineer, experienced in mechanics and optics of motion picture 
cameras, projectors, and film scanning. Give details. Reply to Mr. 
John H. Martin, Columbia Broadcasting System, Inc., 485 Madison 
Ave., New York 22, N.Y. 


Sound recording engineer, 16- or 3 5 -mm equipment, studio or location 
work, single or double system. Free to travel. For details write J. T. K., 
354 Ninth Ave., New York 1, N.Y. 

Honorably discharged veteran with 15 years' experience in all phases of 
motion picture production, including film editing, directing, producing. For 
details write F. A., 30-71 34th St., Long Island City 3, N.Y. Telephone 
AStoria 8-0714. 

Projectionist-newsreel editor with 15 years' experience just released 
from service. Willing to locate anywhere. Write P. O. Box 152, Hamp- 
den Station, Baltimore 11, Maryland. 

Director of visual training aids, now concluding government work, de- 
sires position with educational, industrial or commercial organization, 
supervising or assisting in the production or distribution of visual training 
or allied work. Write H. C. B., 348 Maryland Ave., Dayton 4, Ohio. 

Chief Engineer of motion picture camera manufacturer now available. 
Special training in optics, electricity, electronics, mechanics. Experienced 
in all phases of manufacture of cameras, projectors, and accessories. 
Prefer West Coast, but not essential. Write Robert Winkler, 119 West 
78th St., New York, N. Y. 








Vol45 DECEMBER, 1945 No. 6 



Report of the SMPE Committee on Color 397 

Film The Backbone of Television Programming 


Power Rectifiers for Studio Lighting L. A. UMANSKY 414 

A Small Microphone Boom 


A New Carbon for Increased Light in Studio and 
Theater Projection 


A New Photographic Developer for Picture Negatives 


A Positive Vari-Focal View-Finder for Motion Picture 
Cameras F. G. BACK 466 

Society Announcements 472 

Index of Journal, Vol 45 (July-December, 1945) : 

Author Index 474 

Classified Index 477 

Copyrighted, 1945, by the Society of Motion Picture Engineers, Inc. Permission to republish 
material from the JOURNAL must be obtained in writing from the General Office of the Society. 
The Society is not responsible for statements of authors or contributors. 

Indexes to the semi-annual volumes of the JOURNAL are published in the June and December 
issues. The contents are also indexed in the Industrial Arts Index available in public libraries. 





Board of Editors 





Officers of the Society 
**President: DONALD E. HYNDMAN, 

350 Madison Ave., New York 17. 
** Past-President: HERBERT GRIFFIN, 

133 E. Santa Anita Ave., Burbank, Calif. 
** Executive Vice-President: LORBN L. RYDER, 

5451 Marathon St., Hollywood 38. 

* Engineer ing Vice-President: JOHN A. MAURER, 

37-01 31st St., Long Island City 1, N. Y. 
** Editorial Vice-President: ARTHUR C. DOWNBS, 
Box 6087, Cleveland 1, Ohio. 

* Financial Vice-President: ARTHUR S. DICKINSON 

28 West 44th St., New York 18. 
** Convention Vice-President: WILLIAM C. KUNZMANN, 

Box 6087, Cleveland 1, Ohio. 
^Secretary: E. ALLAN WILLIFORD, 

230 Park Ave., New York 17. 
*Treasurer: M. R. BOYER, 

350 Fifth Ave., New York 1. 


*FRANK E. CARLSON, Nela Park, Cleveland 12, Ohio. 
**JOHN I. CRABTREE, Kodak Park, Rochester 4, N. Y. 
**CHARLES R. DAILY, 5451 Marathon St., Hollywood 38. 

*EDWARD M. HONAN, 6601 Romaine St., Hollywood 38. 
"tCLYDE R. KEITH, 233 Broadway, New York 7. 

*G. T. LORANCE, 92 Gold St., New York 7. 
**PETBR MOLE, 941 N. Sycamore Ave., Hollywood. 
*fHoLLis W. MOYSB, 6656 Santa Monica Blvd., Hollywood. 
**WILLIAM A. MUELLER, 4000 W. Olive Ave., Burbank, Calif. 

*EARL I. SPONABLE, 460 West 54th St., New York 19. 
**REEVE O. STROCK, 111 Eighth Ave., New York 11. 

*WALLACE V. WOLFE, 1016 N. Sycamore St., Hollywood. 

*Term expires December 31, 1945. tChairman, Pacific Coast Section. 
**Term expires December 31, 1946. ^Chairman, Atlantic Coast Section. 

Subscription to nonmembers, $8.00 per annum; to members, $5.00 per annum, included in 
their annual membership dues; single copies, $1.00. A discount on subscription or single copies 
of 15 per cent is allowed to accredited agencies. Order from the Society at address above. 
Published monthly at Easton, Pa., by the Society of Motion Picture Engineers, Inc. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 

General and Editorial Office, Hotel Pennsylvania, New York 1, N. Y. 

Entered as second-class matter January 15, 1930, at the Post Office at Easton, 

Pa., under the Act of March 3, 1879. 


Vol 45 DECEMBER, 1945 No. 6 


As the motion picture industry undertakes an ever-growing per- 
centage of production in color, the matters of increasing present fa- 
cilities, of opening up new facilities, and, above all, the removal of 
limitations and restrictions upon methods of operation, become of the 
utmost importance. The engineers and technicians of the industry 
need to look ahead and formulate the problems and possible solutions 
thereof, so as to be ready wisely to advise the executives and pro- 
ducers, who, it can be confidently predicted, will wake up to the prob- 
lems rather suddenly. 

It is from this point of view that the Color Committee of the So- 
ciety of Motion Picture Engineers wishes to emphasize the importance 
to tne industry of the new high sensitivity caesium-antimony photo- 
tubes which are, to be described in detail in various technical papers at 
this Technical Conference. 

In order to appreciate the significance of these cells it should first 
be realized that the great majority of dyes which can be used in the 
production of color film are transparent to the near infrared region of 
the spectrum and consequently are unsuitable for use as components 
of the sound track, if, as is currently the case, the standard photoelec- 
tric cell has its maximum response in that region. This remark ap- 
plies equally to acid dyes such as are used in imbibition processes, to 
basic dyes as used in dye-toning processes, and to the insoluble dyes 
produced by the aid of color formers and color developers. 

To be sure, there are a very few dyes or suitable pigments which are 
absorptive in the near infrared region ferric- ferrocyanide (the usual 
iron-tone) is the outstanding example. If, then, ferric-ferrocyanide is 
used for the cyan component of the picture, a satisfactory track, ab- 
sorptive in the infrared, can be produced simultaneously with the 
cyan picture component without the requirement of any special proc- 

* Presented Oct. 17, 1945, at the Technical Conference in New York. 



essing steps. Such prints and tracks have been widely used in the 2- 
color field, as, for example, in Cinecolor prints. 

However, the methods which produce this iron-tone image cannot 
in general be used in the production of yellow and magenta images. 
Furthermore, the use of this pigment permits no choice whatever in 
the selection of the cyan component. The iron-toned image for the 
picture has to be accepted as it is. So as a general conclusion, it can 
be said that the picture components suitable for the imbibition proc- 
ess, the color-former processes, or for any process wherein the 3 
components are treated in a common manner, cannot be used to form 
a satisfactory track absorptive to infrared rays. 

This difficulty has been overcome in the past by producing the 
sound track in silver or a silver compound, either by a completely 
separate step, or by means of edge treatment of the film at some stage 
of the processing. As an example of the former method there is 
Technicolor's black silver track, and as an example of the latter there 
is Kodachrome's silver sulphide track. Now either of these methods 
results in added expense in the latter case, because of the delicacy of 
the operation. The boundary of the area of action of the edge-treat- 
ing fluid, whether applied to track alone or to picture alone, must be 
confined to a zone only 0.015 in. wide between track and picture, and 
the action of the fluid must be absolutely uniform across the zone of 
application with no variation in the vicinity of sprocket holes; yet 
agitation as a means of securing uniformity is obviously excluded. 
A delicate operation of this sort requires, on the one hand, careful 
control and, on the other hand, the constant threat of reduced yields 
and increased costs. Quite a number of patents have appeared de- 
scribing various chemical and mechanical methods of performing this 
edge treatment. 

But now, if the sensitivity of the phototube can be confined to the 
visible range, then the same components that are used to make up the 
picture can also be used to produce the track without the necessity of 
edge treatment. It was with this thought in mind that an earlier 
Color Committee (in 1937) called attention to the desirability of find- 
ing or producing such a tube. At about that same time a photosensi- 
tive alloy was discovered by a German investigator, which alloy, when 
illuminated by an incandescent tungsten lamp, has a peak of re- 
sponse well inside the visible range. Furthermore, when properly 
prepared, this alloy possesses very remarkable sensitivity. 

The development of phototubes containing this alloy has appar- 


ently been considerably retarded by the war, though various construc- 
tions containing it have appeared in several countries. German- 
made tubes have been referred to in various articles in Kinotechnik 
and other German magazines, which have been abstracted in the Ko- 
dak Monthly Abstract Bulletin for November, 1943, and February. 
May, October, and December, 1944. A translation of one of these 
articles appeared in the SMPE JOURNAL for September, 1944. Eng- 
lish-made tubes are described in the Proceedings of the Physical So- 
ciety (London) for March, 1943, and also in an article in the Journal 
of the British Kinematograph Society for January, 1943. The U. S. 
patent on the alloy is now in the hands of the Alien Property Custo- 
dian and licenses are available to anyone at a nominal fee. 

For use in projectors it seems most desirable to arrive at a photo- 
tube which will have about the same output as does the ordinary 
Ag-O-Cs tube. A tube which meets this requirement and which is 
mechanically and electrically interchangeable with the present 868 
type has been developed by RCA and will be manufactured in quanti- 
ties as soon as there is a commercial demand. This tube will be known 
as the 1P37. It is to be expected that satisfactory tubes will be made 
available by other manufacturers. 

Tubes of this new type appear to have some advantages over the 
Ag-O-Cs tube. For example, the sensitivity at shorter wavelengths 
makes possible a marked increase in resolution of the slit image. This, 
and related factors, are described in detail in the several papers which 
are scheduled for presentation later in the Conference. From the 
comments of those who have experimented with the new tubes, it ap- 
pears that this new unit is a remarkably good tool and one which prom- 
ises to have a widespread use. In fact, it is very much in order for 
this industry to consider the proposition to replace the present infra- 
red sensitive phototubes in projectors with tubes of this new type. 
The merits of the new unit can be summed up by saying that had it 
been developed prior to the Ag-O-Cs tube there would not now be 
any thought of replacing the former with the latter; and this state- 
ment can be made without considering the new potentialities of the 
new tube for color sound tracks. 

The chief merit of the Ag-O-Cs tube, then, is reduced to the fact that 
it is already in widespread use. But, for sound reproduction, the 
threat of obsolescence hangs over it. 

The results made possible by the use of tubes of this new type with 
sound tracks on color film will be the subject of additional papers at 


this and subsequent Conferences. The way in which tracks of ferric- 
ferrocyanide and of silver sulphide react with the new tube also needs 
study. In this connection it should be noted that ferric-ferrocyanide 
is not nearly so transparent in the range of sensitivity of the new tube 
as are the great majority of dyes in the near infrared region. And it 
appears from preliminary tests that the iron-tone track can be adapted 
to the new tube. 

The Color Committee wishes to stress the fact that in tubes of this 
new type we have a new and useful tool which opens up possibilities 
of simplification of processing of sound tracks on color film. 

Considering the matter of prospective future replacement, the 
Committee wishes to stress the matter of interchangeability and 
standardization, particularly in the 35-mm field. There are ap- 
parently two different characteristic constructions, and it is not clear 
from published data whether or not there is a variation in spectral 
sensitivity between the two constructions. 

If, as, and when new tubes of this description are adopted by the 
industry, such a change must be world- wide and accomplished with 
a minimum of confusion. The Committee recommends that the So- 
ciety communicate with the British Kinematograph Society and 
other similar organizations in all countries, to the end that a norm of 
performance for such tubes be set up. This norm should apply to 
both the optical and electrical characteristics. 

The Committee believes that in the first instance this prospective 
change-over need be considered only for 35-mm film. A change-over 
in the 16-mm field is a rather different matter and with regard to that 
the Committee has no recommendations at the present time. 

Color Committee 

J. A. BALL, Chairman 







Summary. Based upon several years of experience in producing live talent shows, 
facts and figures are presented to show why film and film techniques are better suited 
than live talent for approximately 60 or 70 per cent of the majority of television studio 

Also discussed is the investment in studio equipment and manpower necessary to 
create and transmit, for one time only, a relatively few hours of live talent programs 
per day. 

Twenty-five years ago way back in 1921 radio broadcasting was 
a new, unproved industry. Prominent figures in the advertising, 
manufacturing, motion picture, and newspaper worlds were express- 
ing opinions both pro and con. Some wanted radio broadcasting 
and wan ted it badly; some did not want it at all. Some said it would 
be a great advertising medium ; others said ' 'bosh. ' ' There were some 
who claimed it a wonderful amusement and educational medium. 
Others said "Ridiculous! Who will pay for it?" Some of these men 
knew what they were talking about; some did not. Executives of 
some of the biggest companies issued statements which I am sure they 
would like, in some cases, to have expunged from the record. For 
example, here is a quotation from the Radio Daily dated February 3, 

"A surprising development in radio has arisen with the formation of a committee 
of New York business men to solicit funds from the radio audience of a local sta- 
tion. Money collected will be used to hire entertainers. . . . Word has been re- 
ceived that experiments in radio advertising are to be tried out, to which the de- 
scriptive phrase 'the fourth dimension of advertising' has been applied." 

Here is another dated August 15, 1924: 

"Martin P. Rice, speaking for General Electric, believes that broadcasting will 
eventually be supported by voluntary contributions or by licensing of individual 
radio sets." 

* Presented May 15, 1945, at the Technical Conference in Hollywood. 
** Executive Vice-President, RKO Television Corporation, New York. 


402 R. B. AUSTRIAN Vol 45, No. 6 

And here is another dated December 20, 1924: 

"It is generally supposed that Westinghouse has more than a scientific reason 
for developing the new art of shortwave broadcasting. By its means, they no 
doubt hope to be able to establish chain broadcasting despite A. T. & T. ban on 
leased wires." (Italics by author.) 

Again, under the date of March 13, 1925: 

" 'The radio is not a legitimate medium for advertising; its purpose is to furnish 
people with good music and other forms of entertainment,' said Dr. Lee DeForest 
when he addressed dinner guests of the Harvard Business School Club the other 

Herbert Hoover said at the first radio conference in 1922 : 

"It is inconceivable that we should allow so great a possibility for service, for 
news, for entertainment, for education, and for vital commercial purposes to be 
drowned in advertising chatter ..." 

He made another try at the conference 2 years later in 1924 when 
he said: 

"I believe the quickest way to kill broadcasting would be to use it for direct ad- 

And finally, being very persistent, he said in 1925 : 

"Advertising in the intrusive sense will dull the interest of the listener, and will 
thus defeat the industry. Furthermore, it can bring disaster to the very purpose 
of advertising if it creates resentment to the advertiser." 

Television moves into the scene and we hear the "same record being 
played." Some want it badly; some do not want it at all. Some say 
it would be a great advertising medium; others say "bosh!" Some 
claim it would make a good amusement and educational medium, and 
others say "Ridiculous who will pay for it?" 

I represent a motion picture company. I do not intend to speak 
for other motion picture companies, but RKO has the vision to look 
ahead. We shall, of course, be interested in theater television that 
is television with a box office. We would naturally be interested in 
that. We are interested in home television because some day, ac- 
cording to our own beliefs and the expressions of others whom we be- 
lieve qualified to predict, there may be as many as 30 million television 
screens in the homes of this country and perhaps another 30 million 
screens scattered throughout the rest of the world. If as many as 3 
people look at each of these screens at one time, there is a potential 
audience of 180 million. 

Why should not we be interested? Why should not the advertiser 


be interested in that many screens? I am going to get right into the 
subject of my discussion by making this assumption : I believe tele- 
vision will become one of the greatest mediums for advertising, sales 
promotion, and entertainment we have ever known. I am going to 
try and tell you why I think the motion picture film is the best me- 
dium to convey these messages. 

At the first Annual Conference of the Television Broadcasters As- 
sociation, held in New York in December, 1944, a top executive of a 
large broadcasting network made the statement that television re- 
ceivers would be expensive in the early days of distribution, $300 to 
$350 each and,, to quote him exactly, "Television is not colossal. 
Television is here and the public is ready to receive it when the pro- 
ducers deliver it. My own concept is that no city can support one 
television station unless it has a population of at least 500,000 peo- 
ple." There are in the United States today only 9 cities having a 
population of a half million or over. At the present time there are on 
file with the Federal Communications Commission applications from 
hard-headed businessmen for nearly 120 television stations in about 
45 different cities from 45,000 population up. 

I would like to quote from testimony given before the FCC fre- 
quency allocation hearings in October, 1944, by an executive of one 
of the largest networks, NBC in fact. This gentleman is charged with 
the duty of formulating plans for the development and expansion of 
television networks. "During the course of this hearing, the Commis- 
sion has indicated its interest in how far the service of television can 
be made available to those living in nonmetropolitan areas. My stud- 
ies and discussions with affiliates have convinced me that television 
broadcasting is practical in a town of 25,000 with normal density of 
population in the surrounding rural areas." Now-someone is wrong! 

I quote further from the first executive's statements : "I further be- 
lieve that not more than 10 per cent of the people in the foreseeable 
future are going to be eligible to receive television receivers either 
from the viewpoint of income or the ability to get satisfactory recep- 
tion from one or more television stations in their community." This 
utterly amazes me. I am sure all of us can remember how we paid 
what now appears to be exorbitant prices for crude radio sets with 
their large collection of external batteries, chargers, loudspeakers, etc. 
I do not know of a single manufacturer who does not feel confident 
that he will put out a good, workable television receiver for as low as 
S175-$200, and perhaps even lower. I cannot but feel that this gentle- 

404 R. B. AUSTRIAN Vol 45, No. 6 

man's views are based not on facts as much as they are based on desires 
and personal preferences. The same speaker also said that films, for 
a while, may supply part of the programming for television and for a 
limited time the public would tolerate films. "* 'Tolerate" is his word. 
May I humbly call your attention to the fact that over 80 million 
people a week go to the movies and pay for it? The annual box-office 
"take" in 1943 was $1,800,000,000 and it was higher in 1944. I think, 
therefore, that if we give people good entertainment on film for noth- 
ing, they would not only "tolerate" it but would welcome it in com- 
parison with some of the live talent shows it has been my misfortune 
to see. 

Here is another statement this time, by a very good friend of 
mine, Gilbert Seldes, Director of Television Programs, Columbia 
Broadcasting System. In an article written for the New York Times 
dated December 24, 1944, he refers to a little live talent dramatic 
sketch he put on and of some of the problems incidental thereto. 
"We found that all our estimates for rehearsal time with televisors 
were grossly low. Much can be done in advance ; but you have to go 
under the lights and see the results in your monitors before you know 
what you've got. We used, actually, some 8 different sets in 30 min; 
they ranged from a blank wall a neutral shot against which the 
narrator began and ended his story to a kitchen in a cabin, complete 
with stove. For the most part we managed to keep the action flowing 
steadily; when changes were too complex to be made in a few seconds, 
we added 10 or 15 sec by tnrowing on the screen a symbolic image, the 
outside of the cabin, or a bus sign, or a recruiting poster. We felt 
acutely the need for speedy movement. We recorded certain long solo 
speeches. Among other advantages, the records allowed us intervals 
of a minute or two for making set changes." 

It seems Mr. Seldes had to use every device he knew of to pad and 
slow down the performance, which is deadly in showmanship. We in 
the movies think of every device we possibly can use to speed up ac- 
tion, not slow it down. Had Mr. Seldes been working with films, he 
would have had no problems that a pair of scissors could not cure. 

He now speaks in this same article about new techniques that tele- 
vision brings. "The moments when you hit this special quality are 
the biggest kick in television right now, and your audience gets it, too. 
The other day we got it in another experiment; we took a series of 
drawings of a WAC giving jiu-jitsu to an apache dancer." (Now 
there is a program that must have been thrilling! That is going to 


make television!) Mr. Seldes said, "We got 2 dancers to re-enact the 
strip, and as they reached certain high spots, we dissolved from them 
to the appropriate section of the original strip of drawings." (Now, 
get this sage observation ) "It was not an esthetic discovery; it 
was only something that could not be done anywhere except in tele- 
vision. It had a unique quality. And our audience responded to it 
immediately." I frankly do not know what Seldes means when he 
then says, speaking of this latter invention, "It can only be done in 
television." It was done 30 years ago, in the movies. 

It was because of observations such as these and others that I 
opened my address in the Theater Panel of the Conference of the 
Television Broadcasters Association with the following statement, and 
I wish to reiterate it: "There has been some unfavorable comment 
concerning the use of film in television programming. Why? Some- 
times, because of prejudices, desires, wishful thinking, unfamiliarity 
with the possibilities of the medium, inability to use it correctly, or 
just plain ignorance." 

The use of the motion picture for the presentation of a television 
program immediately frees the writer, director, and producer from 
the shackles of the live stage. A television program using live talent 
naturally faces many of the same restrictions that limit a stage pro- 
duction. All action at any given time is on a single set, or a group of 
simulated sets, and because the action is continuous, the players are 
held to a single costume. Offstage action can only be referred to, and 
not shown, unless an intermission is declared for a shift of scene and a 
change of costumes. But with television there can be no between- 
the-acts intermission. An intermission on a television program would 
cause the audience to dial promptly to another station. True, a sys- 
tem of revolving stage sets, multiple cameras, and quick costume 
changes might be employed, but the technical complications would 
be enormous. At best, it would permit only a very few locale changes, 
but would continue to impose a multitude of restrictions on both 
writer and producer. There can be no reverse shots for, naturally, 
there is no stopping the show and moving cameras and gear to a new 
setup. Live talent proponents are talking of as many as 6 cameras 
on one set at one time. I am wondering how they propose lighting 
such a set. How do they propose to light the actors so they even ap- 
proach the present motion picture lighting standards? I hesitate to 
even estimate the rehearsal time necessary for such an elaborate set- 

406 R. B. AUSTRIAN Vol 45, No. 6 

All of these handicaps and restrictions, however, are immediately 
removed when it is a film program. As for locale changes, the problem 
is most simple. Should the script call for an authentic street scene 
in Calcutta or Miami, a blizzard in Alaska, or a storm at sea, it would 
be available from the extensive film libraries maintained by RKO in 
Hollywood or Pathe News in New York. The players called for in 
the script would perform against the background of such a scene in 
such a natural manner that to the television viewer it would all appear 
to be taken on location. When film is used, rather than live talent, 
this and hundreds of other proved motion picture devices can be em- 
ployed to give the television program producer practically limitless 
freedom of action. 

One of the ever-present dangers in live talent production is the 
"fluff" or "blow up." This will ruin any dramatic performance on 
television. A stage wait just a second can snap the emotional 
thread of any show. A television director must cut his show as it 
goes on the air push a wrong button fade in the wrong scene dis- 
aster. There has never been any substitute for pre-editing. Cer- 
tainly no sane advertiser would trust to any demonstration of his 
product before millions of pairs of eyes if there was the barest possi- 
bility of a slip. I actually saw a piece of nonbreakable glass, bearing 
a nationally known trademark, break into a million pieces when it 
was subjected to a hammer test. An incident like that could laugh a 
10-million dollar company out of business over night. 

Another great plus in the use of film is the availability of the ani- 
mated drawing, either separately or in combination with regular film 
presentations. The possibilities of animation for the advertiser in dia- 
grammatically demonstrating his product also opens up a fascinating 
new world for commercials. 

I believe that the most suitable type of television programs, as they 
are evolved through experience, will be far different in character from 
the feature motion pictures created in Hollywood for theater exhibi- 
tion. No hour and a half and 2-hr shows. Fifteen- and 30-min 
shows, even shorter. In the evolving of such programs, however, 
motion picture techniques will play a greater part than existing radio 
techniques and the use of film will be infinitely more important in tele- 
vision broadcasting than the electrical transcription disk is today in 

Another objection to programs on film has been that they do not 
have the sense of "immediacy" that live talent has. I do not think 


that is true except, of course, in the case of sporting events, spot news, 
fires, floods, etc. There is no one who appreciates the live performance 
provided by the legitimate stage any more than I do. There is some- 
thing about sitting in a theater watching a good play performed by 
capable actors and actresses that has no substitute; but if you were 
to put television cameras in that theater and watch those actors and 
actresses from a distance, it is then nothing more than a motion pic- 
ture, and because of its cramped locale and immobile cameras, a bad 
one! It loses the personal magnetism, the feeling of live flesh and 
blood that thing known as "theater." 

Here is another reason why I cannot agree to the theory of imme- 
diacy. Every day as I come to my office, which is in the same build- 
ing as the Radio City Music Hall, I see literally hundreds of people on 
line in all kinds of weather patiently waiting to go inside and see a 
picture which they know was completed several months ago. The 
people know the middle scenes were shot first and the first scenes shot 
last. They know each scene was taken and retaken. Yet they line 
up to get in. Why? It is only a 2-dimensional moving shadow, but 
the actors and actresses will make them laugh or cry, or forget them- 
selves for a few fleeting hours. It is showmanship! 

Television, like motion pictures, is showmanship by remote con- 
trol. It has been said many times : The high cost of programs on film 
is an impossible obstacle. I do not believe that programs, in order to 
be entertaining and good, necessarily have to be expensive beyond 
reason. Some radio programs today cost from $10,000 to $25,000 for 
a 30-min period. That is a range of from $300 to $600 a min. We 
know we can supply film shows for that much and less and, of course, 

I am worried about the high cost of live talent programs the vast 
amount of equipment and personnel needed to put a comparatively 
few hours of live talent programs on the air. This is the knotty little 
economic problem of television which keeps occupying the attention 
of prospective station operators, especially the so-called originating 
stations for networks. There are a lot of people who say that tele- 
vision will operate only between 4 and 6 hr a day. Others point 
knowingly to a 24-hr-around-the-clock schedule. Let us take for our 
example a day which calls for 8 hr of live talent studio programs. 
What would be required to put on 8 hr of programs a day if they were 
ah* live talent programs? 

Thomas H. Hutchinson, Director of Production of RKO Television 



Vol 45, No. 6 

Corporation, has prepared a chart, shown as Fig. 1, and supporting 
explanation : 

Charf efsuoqesfedsfvaf/o operaf/ons -for e/'o/rf- hours per cfoy 
of /V 7e/ev/siofr Programs. 


j j 1 ! ! J } 

8 HOURS AIR TIME. \ 30%30"i 30"*30" \ 15*30430'! 30*30:30'; 30%30" ;I5'130'*30'; 30*^30" 

53 HRS. 15" REHEARSAL \ 6hrs.30"! 6 hrs. ; 8hrs.l5"| 9hrs.30"| 8 hrs. j 9 hrs. j 6 hrs. 




Technical Director 

Video Engineer 


Sound Effects operato 

Camera Man 
Camera > 
Dolly " 
Asst. .- 

Property Man 
Boom Sound 
Sto^e Manager 
Video Effects 

I8hr$. 15" Scenery Crew 2 hrs. 2hrs. 3 hrs. 
Stwdio Operatinq Crew 8 hrs. 8 hrs. IOhrs.30" 

3 hrs. 2hrs.30" 3hrs.45" 2 hrs. 
12 hrs. 10 hcs. lllirs.30" 7hr$.30" 

Tliis chart is based on the assumption that each sfudi'o is of sufficient- 
size to allow scenery for two half hour proqrams to be set up at one time 
C Three cameras, two of them on dollies would be used.} 


FIG. 1. 

"The chart of studio operations for 8 hr a day is based on several very important 
assumptions. First, that each studio is of sufficient size to allow for the placing of 
all the scenery to be used in 2 half -hour television programs. Secondly, it is as- 
sumed that each 30-min unit can be rehearsed in approximately 3 hr, before the 
cameras. No consideration has been given for previous rehearsals. In studio 1 


there are shown 5*/2 consecutive hr of rehearsal before the first program goes on 
the air. 

"It is presupposed in this chart that the second program will have been re- 
hearsed during part of this 5V2-hr rehearsal period. The program to be broadcast 
between 4:30 and 5:00 might rehearse from 8:00 to 10:00 in the morning and 
from 3 : 30 to 4 : 30 in the afternoon. The first program from this studio scheduled 
from 2:30 to 3:00 would rehearse from 10:00 to 12:00 and from 1:00 to 2:30. 
This would allow 3 hr rehearsal for the program broadcast from 4:30 to 5:00, and 
3 X A hr for the program broadcast between 2 : 30 and 3 : 00. 

"No attempt has been made to indicate the exact number of studio operating 
crews necessary in a day's cycle. In studio 4, for example, there are shown three 30- 
min programs. If the procedure suggested in studio 1 were applied here, one oper- 
ating crew would report at 8 A.M. and rehearse until 11 :30. The program would 
then be broadcast from 11 : 30 to 12:00. At 1 o'clock a new operating crew would 
report and would rehearse and broadcast the 2 programs scheduled from 7:00 
to 7:30, and from 9:30 to 10:00. No attempt has been made to fill in the extra 
day for the first operating crew. Under this procedure they have only worked a 
4-hr day, but at the same time, unless there was an additional studio available, 
there is no way with only 7 studios to utilize their services in active operation for 
the other 4 hr. 

"It will always be necessary to balance operating man-hours against studio fa- 
cilities. It will always be virtually impossible to get 100 per cent efficiency out of 
any operating crew. Time for changing scenery during rehearsal periods has been 
arbitrarily allotted. This has been done in an attempt to increase the efficiency of 
the scenery crew. It will readily be seen that if the scenery time scheduled on the 
chart during an operational day is eliminated and all scenery is changed after the 
studio is through for the day, it will take a scenery changing crew for practically 
every studio, whereas, if it is handled as suggested in the chart, the number of 
crews necessary may be considerably reduced. 

"This chart is also based on the assumption that the programs to be broadcast 
can be rehearsed satisfactorily on a 6-to-l ratio. All programs will not take this 
long and proper planning should make this schedule workable, but programs may 
have to be changed to conform to the rehearsal time available. No reference has 
been made in this chart to the number of men involved in the scenery crew. That 
will depend entirely on the amount of scenery used for each program. No refer- 
ence has been made as to the amount of scenery that will be required. This will 
depend entirely on the type of programs produced and the production standards 
set by the station. It seems only reasonable to assume that the 17 program units 
indicated on the chart would require at least one set each, and some programs 
might run as high as 5 or 6 sets per unit. All of these sets must be designed, built, 
and painted and no attempt has been made on this chart to estimate the amount of 
time or the number of men required to do this. 

"If the schedule were to be changed so that one consecutive hour of program was 
presented rather than two 30-min programs, it might very well be that much more 
rehearsal time would be necessary than is indicated here. 

"Roughly, it can readily be seen that each hour of program per day is going to 
require a studio, an operating crew of 18 men, a crew to change scenery, and a crew 
to build and paint it, but the chart does show how one studio might be eliminated 

410 R. B. AUSTRIAN Vol 45, No. 6 

in a normal day's operation if the programs were planned and produced as indi- 
cated. The chart only covers production for one day. On a 7-day operating week 
it is obvious that it would be necessary to use swing crews for the extra 2 days as 
the same crews should produce the same programs each week." 

How very simple it is going to be when programs are provided on 
film. A large motion picture company can arrange with the advertis- 
ing agency for whom it functions to supply identical prints of a pro- 
gram to any number of television stations throughout the country or, 
for that matter, the world. It would do this precisely as it supplies 
its regular motion picture film, on a "day and date" basis. This would 
be done through its nation-wide network of film exchanges. No ex- 
citement, no worry, no scurry just as simple as loading a home movie. 

Public relations via television opens up a vast new field of oppor- 
tunity. Today, more than ever before, it has become necessary for 
big business to justify its existence. Television on film will offer a 
most unique and effective method of spreading the story of the large 
corporation to the public. In a most entertaining manner it will be 
possible to portray what a big company does for its employees 
group insurance, social service, hospitalization, home economics, 
company stores, extension courses, bonus system, job insurance, re- 
tirement funds. What better way could there be to present this story 
to the peoples of the world than via the motion picture films? 

The screens of the majority of motion picture theaters in this and 
many other countries have been closed to the advertising or business 
propaganda film, and rightfully so. People do not wish to buy 
propaganda or advertising when they- go to the theater. Television 
will open the home screens of the world to the advertiser. Here 
again the efficacy of the television program on film becomes apparent. 
You cannot drag television cameras all through a big plant and put 
on a carefully planned show. Also the finest live talent program pro- 
duced in America means nothing in a country where English is not 
spoken or understood. The cost of the program or series of programs 
must be borne by the one-shot performance. Suppose it were to cost 
10 times as much to put it on film? By the simple expedient of "lip 
dubbing," we can translate the program into any and every language 
and send prints of such programs to all countries just as we now send 
foreign versions of our motion pictures. This enables our giant cor- 
porations whose scope is world wide to reach via television not just 
millions in this country but, eventually, billions of people. 

The cost of installation and operation of a television station will be 


considerably higher than that of a radio station. However, a mes- 
sage or an advertisement can definitely be put across in much less 
time than it takes via radio. That will offset the higher operating 
costs. Radio programs are sold and produced in 15-min periods. 
Why not 10-min periods ? There would be a 33 Y 3 per cent saving right 
there. The show would probably be tighter and faster moving, too. 

I hope it will always be remembered that television is primarily a 
visual art. I hope that it will not be burdened with the top heavy 
verbosity we hear every day over our radios. I do not think that the 
American public is quite so dense or impervious to suggestions as some 
of the radio commercials we hear would lead us to believe. Suppose 
any one of you were this evening to stroll through the lobby of this 
hotel. There you spied a very attractive young lady. You ca tight 
her eye and she caught yours. She looked right at you, smiled, and 
perhaps gave you a knowing little wink. I wonder if there is any- 
one in this room who would not know what to do in such a case or 
whatnot to do! I cannot conceive of having a hearty- voiced an- 
nouncer from somewhere give you a 2-min sales talk ! 

In almost every motion picture you see, there are several complete 
situations some of them involved, which are conveyed to the be- 
holder with complete clarity in a twinkling of an eye without a spoken 
word. One of these methods is known as the "double take." You 
have ah 1 seen it used. A quick look, and then another look. A gesture, 
a movement of the hand, an expression of the face, a sly wink. Each 
of these devices can convey more immediate understanding than a 
hundred words. Barry Fitzgerald feels that in contemporary pictures 
there is too much dialogue. The finest acting, he says, is still panto- 
mine. A look, a twist of the neck, or the way a person walks, can 
tell more than whole pages of dialogue. In Going My Way, for in- 
stance, the glance of disapproval he gave his curate, Bing Crosby, 
when he discovered him wearing a sweater with "St. Louis Browns" 
spelled on it, or his silent shame when he learns that the turkey on which 
he is feasting has been stolen, tells more in less time than even the 
brightest dialogue. Good television will definitely shorten the amount 
of airtime needed to put over a message. Let us hope and pray we 
can so convince the advertising agencies. 

During the past few months many inquiries from prospective ad- 
vertisers and prospective telecasting station operators have been made 
which indicate that while they have great faith in the ultimate future 
of television, they are rather puzzled as to how they can program their 

412 R. B. AUSTRIAN Vol 45, No. 6 

stations. Sooner or later the discussion veers to the use of programs 
on film, film transcriptions, or, as RKO refers to them, Telereels. 
The average radio station operator, the word "average"' here meaning 
one located at a distance from one of the major metropolitan districts, 
is now dependent upon programs that reach him from talent centers. 
Ordinary radio broadcasting has shown us that local talent cannot 
supply more than a fraction of the needed program material. 

It is quite evident that telecasting stations will be in operation 
considerably sooner than network programs will become available 
and it is here that the film transcription or Telereel will prove to be 
the backbone of the programming system. 

RKO is going to make syndicated programs available as soon as 
station construction starts. Even after national or large regional net- 
works are established, the Telereel will remain an important, if not 
the most important, factor for all programs with the exception, of 
course, of sporting events and news events which are always hot, 
flash news. 

No single individual advertiser, no single advertising agency, nor any 
group of advertising agencies could possibly operate such enormous fa- 
cilities as RKO and its subsidiary, Pathe News, Inc., now offer the po- 
tential television users of this country. These facilities are now available 
to both reputable advertisers and recognized advertising agencies 
through RKO Television Corporation. 

In the post-war period when television will flourish, the advertising 
dollar will be scrutinized more carefully than it is today. National 
advertisers will not be so ready to buy a 15-min or half -hour spot on a 
network between the East and West Coasts and perhaps be in compe- 
tition with a top rating program carried by another network, or face a 
3-hr time differential. Advertising managers and market analysts 
will lean heavily on the spot type of telecasting. They will pick the 
markets in the order of their desirability, concentrate their appropri- 
ations on selected territories, and make it a point to cover them at the 
best possible hours. Perhaps they might even give a repeat show. 
Why are good radio programs today not repeated? We can only 
listen to one program at a time. How often has a friend of yours said 
to you one morning, "Did you hear so-and-so on the radio last night? 
He was marvelous." You say, "No, I didn't. I was listening to an- 
other program, or, I was out last night. I'd like a chance to hear that 
program you speak of it sounds wonderful." I do not think adver- 
tisers today get nearly the circulation from their radio programs they 


should because of this one-shot feature. A Hooper rating of twenty 
says to me, "What happened to the other eighty?" 

I believe it will be perfectly possible and feasible to release a pro- 
gram over a "first-run" group of stations, and then rerelease it in the 
same locality at a later time to a "second-run" group, and finally a 
"subsequent-run" group. Any given locality can be thoroughly and 
completely covered. The Hooper rating will go well up and the "cost 
per thousand" listeners go way down. The television industry can 
learn a lot from the motion picture industry. 

Obvious, indeed, would be the great saving in land line <or radio re- 
lay charges by the use of Telereels, and obvious, too, is how much 
greater a percentage of its rate card a local station operator would be 
able to retain. 

I do not wish to create the impression that it is my belief that there 
will not be any live programs. This is farthest from my mind. There 
are certain types of programs that will always be done best in the 
flesh, as it were, interviews of prominent people, style and fashion dis- 
plays, all kinds of sporting events, outdoor pageants, in fact, any 
event whose main attraction is uncertainty of the outcome, such as a 
football game. But, as you have undoubtedly gathered from my re- 
marks, I am very "bullish" regarding the eventual triumph of film 
over live talent. It will be the backbone of television programming. 

This is my case, then, for the employment of the motion picture to 
carry the public relations message, advertising, education, and en- 
tertainment via television. I recommend the employment of the 
same medium that has so successfully spread American culture and 
American ideas over the face of the entire globe. 



Summary. Mercury arc rectifiers are now widely used in the industry for 
converting a-c to d-c power. Since d-c power is required for operation of arc lights 
in motion picture studios, it is of interest to analyze whether the use of rectifiers for 
this purpose has advantages over other power conversion means. This paper points 
out the relative merits of rectifiers and motor generators. 


Motion picture studios, employing a great number of high-intensity 
arc lights, require a large amount of d-c power. In some cases as 
many as 8, 10, or more motor-generator sets, up to 500 kw capacity 
each, are employed in a single substation supplying direct current to 
several studios on the same location. 

The d-c power system is usually arranged as a 3- wire, 1 25-250- v 
circuit, in order to save on distribution copper, with arc lights con- 
nected between the neutral and either polarity. 

Mercury arc rectifiers have firmly established themselves in many 
industries as the preferred means of a-c to d-c p6wer conversion. It 
will not be amiss to state on this occasion that during the peak year 
of war production about 10 per cent of all electrical energy produced 
in the United States has passed through rectifiers and has been utilized 
as direct current. Close to 3,500,000 kw of rectifiers are installed in 
this country alone. While the bulk of this capacity is provided for 
the electrochemical plants, there is hardly any industry where the 
rectifiers have not proved themselves as a reliable, and often pre- 
ferred, conversion means. 

It is, therefore, pertinent to review the possibilities of the use of 
rectifiers for the application in which we are interested, i. e., for 
supplying d-c power to the studio lights. 

* Presented May 18, 1945, at the Technical Conference in Hollywood. 
** Industrial Engineering Divisions, General Electric Company, Schenectady, 
N. Y. 



To provide the proper background for our discussion, a bird's-eye 
review of many typical rectifier applications is in order. 

FIG. 1 . A 3000-kw, 650-v rectifier supplying power to New York'City subways . 

FIG. 2. 3000-kw, 3000-v rectifiers supplying power to DL&WRR. 

The first rectifiers in this country were used for electric railways 
and street cars. In the first place, the rectifier, being a static device 
with practically no moving parts, permitted the design and building 



Vol 45, No. 6 

of unattended a-c to d-c substations quite an item in operating ex- 
pense of a railroad. Secondly, the high d-c voltage 600 v and 
higher used for this purpose, particularly favored the rectifier since 
its efficiency goes up with d-c voltage, and its first cost goes down. 

For instance, Fig. 1 illustrates a 3000-kw, 650-v rectifier installed 
in New York City subways. A total of 261,000 kw of rectifiers were 
installed on this system since 1930. Fig. 2 shows a group of four 
3000-kw, 3000-v rectifiers supplying power to the suburban line of the 
DL&W Railroad. 

FIG. 3. Twelve rectifiers supplying a total of 60,000 amp at 650 v to an 
aluminum electrolytic pot line. 

Since just before the war, the use of rectifiers in the electrochemical 
industry began to grow by leaps and bounds. In the production of 
many chemicals, and particularly in the production of aluminum and 
magnesium, an electrolytic line is part of the process, and a very im- 
portant part at that. Many so-called "pot-lines" for these light 
metals require 50-60,000 amp each of direct current at voltages of 
about 650 v, more or less. Fig. 3 gives a general view of such an in- 
stallation involving twelve 5000-amp, 650-v rectifiers in an aluminum 
plant. Over 2,500,000 kw of rectifiers have been installed within the 
last few years in the chemical industry alone, to the exclusion of any 
new rotating machines as a-c to d-c converting means. There are a 
good many reasons for this preference. 



Efficiency is higher (see Fig. 4) while the cost of power is an im- 
portant item in the total production cost ; for instance, even at 2 mils 
per kwh, a pound of aluminum will require 2 cents worth of electric 
power as compared with the selling price of 15 cents per Ib. In- 
stallation costs with rectifiers are lower. Maintenance expenses are 
reduced. Fewer, if any, substation attendants are needed. These 
and other factors have decided the issue in favor of rectifiers in that 
industry as in many others. 




> 90 










40 60 80 100 

FIG. 4. Efficiency of a 3000-kw, 600-v rectifier as com- 
pared with rotating conversion equipment. 

Fig. 5 shows a 1500-kw, 250- v rectifier used in a steel mill to supply 
general purpose power. Here, again, the substation can be made un- 
attended. While the full load efficiency of the rectifier even at this 
voltage is somewhat higher than that of a motor generator, the im- 
provement of the part-load efficiency is particularly noteworthy (see 
Fig. 6) ; in this type of service the load is intermittent, and the over- 
all load factor is not always very high. Therefore, the all-day ef- 
ficiency of the rectifier substantially is higher than with a synchronous 
motor-generator set. 

Many other diversified industries took to rectifiers as well. For 



Vol 45, No. 6 

capacities up to 500 kw, 250 v, which are usually required in many 
such cases, an enclosed design (Fig. 7) has proved to be quite popular. 
Incoming line a-c switchgear controls, the primary of the rectifier 
transformer to which it is "throat-connected." The transformer is 
likewise connected to the rectifier cubicle and d-c switchgear. Fig. 8 
shows the inside of the rectifier; it includes several steel- jacketed, 
water-cooled ignitron tubes. More will be said of these tubes later 

FIG. 5. A 1500-kw, 250-v rectifier used in a steel mill. 

Rectifiers are also used extensively underground in mines. In this 
case they are usually made portable, as shown on Fig. 9. The first 
car carries the primary a-c switchgear, the second car, the transformer, 
and the third car, the rectifier tubes, d-c switchgear and accessories. 
As the mine is being worked farther and farther, the rectifier is also 
moved to be nearer the center of the load. 

This brief review gives us a general picture of the place taken by 
the rectifiers in industry. Our next step is to have a closer look as to 
how the rectifiers are built, and what makes them work. 


The heart of each rectifier is, of course, the electronic tubes which 
actually convert the alternating current into direct current. 

Dec., 1945 



High vacuum or "hard" tubes are not ordinarily used for power 
rectifier work for obtaining commercially usable d-c voltages and 
currents. Gas filled, hot cathode tubes, such as phanotrons and 
thyratrons are employed for moderate capacity up to about 25 amp 
per tube. For larger capacities, such as we are interested in, the 
mercury pool tubes are almost universally used. 





iu 86 




tr 82 















FIG. 6. Efficiency of a 1500-kw, 250-v rectifier as com- 
pared with a motor generator. 

The most popular of these tubes is the ignitron, shown in Fig. 10. 
It is a metal envelope, water-cooled tube, with a graphite anode, and 
a pool of mercury acting as cathode. Using mercury for this purpose 
gives us several advantages. 

The electrons of the mercury atoms are loosely held by the positive 
charge; a lower temperature and lower voltage are required to emit 
electrons than would be the case for other metals. Mercury vapor- 
ized by the cathode also offers a means for production of electrons in 
its own vapor. The recombination of electrons and ions constantly 
occurs at the same rate of ionization. These mercury atoms recon- 
dense and return to the cathode pool. Therefore this pool is con- 



Vol 45, No. 6 

tinually and automatically maintained. "The mercury does not 
wear out." 

An ignitron tube is "started" for each cycle, by creating a new 
"cathode spot" for each cycle. This is done by means of an igniter 
whose point, made of boron carbide, is dipped in the mercury pool. 
At a certain instant of each cycle a voltage is applied to the igniter 
from a special "peaking" transformer, and current flows through the 
igniter to the mercury. Since the mercury does not wet the boron 
carbide, the contact resistance is rather high and enough heat is 

FIG. 7. Front view of a metal enclosed sealed tube rectifier, 300-kw, 250-v, 
typical of many similar units used in various industries. 

generated to produce the cathode spot. The tube becomes conduc- 
tive and carries current for the rest of the half -cycle. Like any 
other electronic tube, the ignitron does not conduct any current for 
the negative half -cycle; i. e., when the anode is negative in respect to 
the cathode. 

Therefore, the igniter point should be energized or "fired" every 
cycle very much as a cylinder of a gas engine is fired periodically. As 
will be shown later, the firing circuit is derived from the same source 
of power as the rectifier itself, i. e., complete synchronism is main- 
tained, just like the timer and distributor of an engine are driven in 
synchronism with the engine itself. The analogy may be drawn still 




further: by retarding or advancing the "spark" or ignition, we can 
modify the output of the rectifier. 

Ignitron tubes, as shown in Fig. 10, are sealed off at the factory, and 
are now available in continuous ratings of 100 amp and 200 amp at 
d-c voltages of 250 v or less. Their current rating is somewhat re- 
duced at higher d-c voltages, like 600 v. By properly combining 
these tubes, as we have already seen, rectifiers up to 500 kw, 250 v, 
cr 1000 kw, 600 v can be readily provided. 

For larger capacities the so-called pumped ignitron rectifiers are 
widely employed. Fig. 11 illus- 
trates the cross section of such *****^- _ __ LMIEMIIIMI^^^BI 

tube or tank. It consists essen- 
tially of the same components as 
the sealed ignitron. However, 
each unit is connected through a H 
vacuum tight valve to a mani- 
fold which is continuously evacu- 
ated by a pump, shown in Fig. 
12. This photograph represents 
an assembly of a 4000-amp recti- 

A complete vacuum pumping 
system consists of 2 vacuum 
pumps connected in series, which 
usually operate continuously 

when the rectifier is in service. 

~, - . FIG. 8. Inside view of sealed tube rec- 

The primary or roughing pump tifier> with compartm ent doors open. 

is a motor-driven oil immersed 

vertical-type compressor similar to units used in refrigerators. It is 
connected in series through an expansion tank to a mercury condensa- 
tion pump. In this manner, an almost perfect vacuum of 1.0 micron 
(1/760,000 part of an atmosphere pressure), or less is maintained. 

A pumped rectifier can be dismantled in the field, if desired, and 
then reassembled, evacuated (or degassed), and put back in service. 
The igniter point can be readily changed. Thus, a pumped rectifier 
has no inherent life limit. 

There is a definite arc drop voltage in the rectifier tube, even 
though the advent of the single anode tubes, such as ignitrons, has 
materially reduced this value. 

In the smaller ignitrons of the sealed type this arc drop is from. 1 5 

422 L. A. UMANSKY Vol 45, No. 6 

to 19 v. It is somewhat higher in the larger, pumped equipments. 
The arc drop results in power loss in the tube, which is converted into 
heat and is removed by cooling water. 

The value of the arc drop does not depend on the voltage at which 
the tubes are operating. It is obvious, therefore, that the rectifier 
is at its best, as far as its efficiency is concerned, when operating at 
the maximum rated voltage; better at 600 v than at 250 v; better at 
250 v than at 125 v. 


With the present interest in "things electronic," the majority of 
engineers have a general understanding of the performance of simple 

FIG. 9. A portable mining rectifier substation, rated 300 kw, 275 v, mounted 
on 3 flat cars and designed for low headroom. 

electronic circuits. Some of the diagrams are, undoubtedly, familiar 
to many, but they are included for the sake of completeness and as a 
ready reference. 

For capacities in which we are interested, we are always assuming 
a 3-phase a-c power supply. Fig. 13 (a) gives the simplest arrange- 
ment with 3 tubes used for rectification. The anode which, at any 
given time, is the most positive carries the entire load current. In 
this case it carries it for 1/3 cycle (120 degrees), after which the cur- 
rent is transferred to the next anode. This transfer is called "com- 

The resultant d-c voltage is illustrated by the heavy curve, Fig. 
13(b). The ratio of average d-c voltage to maximum values is 
0.825; the "ripple" is quite pronounced. Therefore, the 3-phase 



power rectifiers are not as frequently used as those with a larger num- 
ber of phases. 

Fig. 14 shows a 6-phase (diametrical) arrangement of tubes and 
transformer secondary winding. The ripple is greatly minimized. 
The ratio of average d-c voltage to maximum values has risen to 0.97. 
Each tube carries the load current for only 1/6 cycle. 












\ OFF." 

FIG. 10. 

Cross section of a sealed ignitron tube used for power 
rectifier work. 

The action of transferring current from one anode or phase to an- 
other anode is similar to commutation from coil to coil of a d-c ma- 
chine. This commutation is not instantaneous because the reactance 
of the several windings involved resists any sudden transfer of cur- 
rent. Therefore, the actual shape of voltage and current is repre- 
sented by Fig. 14(c), differing from the theoretical shape shown in 
Fig. 14(b). This means that for a short time during the transfer, 
current flows simultaneously through 2 anodes. 

The "notches" in the voltage curve, Fig. 14(c), show that the 
average d-c voltage is reduced owing to the commutation factor. 
This reduction is proportional to reactance of the circuit and to the 



Vol 45, No. 6 



















FIG. 11. Cross section of a. pumped ignitron tank used for power rectifiers of 

larger capacity. 


value of load. This explains the well-known fact that with a-c 
voltage being maintained constant, the rectifier d-c voltage will have 
an inherent regulation of about 6-10 per cent from full load to no 

Fig. 15 shows the most popular connection of rectifiers used in the 
great majority of industrial applications; in fact, probably better 
than 90 per cent of all rectifiers are so arranged. Considering the 

FIG. 12. Shop view of a pump rectifier, 1000 kw, 250 v, 
showing the vacuum pump. 

transformer windings, the arrangement is known as "Delta, 6-phase, 
double wye." 

Comparing Figs. 14 and 15, we note that the wave shape of d-c 
output voltage from these 2 circuits is the same, the fundamental 
ripple being 6 times the basic frequency (i. e., 360 cycles with 60- 
cycle supply). The fundamental difference lies in the action of the 
interphase transformer, which is the midtap reactor connected be- 
tween the 2 wyes. 

Each 3-phase group or wye operates as a 3-phase rectifier. The 
wye-points of both groups are interconnected to the so-called "inter- 
phase transformer." The latter is usually mounted in the same tank 



Vol 45, No. 6 

or enclosure as the main transformer, but is wound on a separate core, 
and is magnetically independent. 

Assume now that at some instant the phase (1) gives its anode the 
highest positive potential. The current then flows through the tube 
No. 1, the cathode (positive) bus, the load, and returns through the 
midpoint (7) of the interphase and to the neutral of the 3-phase 
group 1-3-5. As a result, the effective voltage of phase (1) is lowered 
by the amount of the impedance drop across one half of the interphase 
transformer. The passing of this current induces voltage in the 
other half of the interphase and, in effect, raises the voltage of phase 


FIG. 13. Three-phase rectifier circuit (a) and the wave shape 
of d-c voltage (6). This circuit is used for relatively small 
power rectifiers. 

(2) which is next in the line to assume the carrying of the load. The 
result of decreasing the potential of (1) and increasing that of (2) is 
that for an interval of 60 electrical degrees; the anodes (1) and (2) 
are at equal positive potential which is then higher than that of any 
other anode in the circuit. Therefore, for this length of time the 2 
anodes share the load; during the next Ve cycle the load is carried 
by anodes (2) and (3) ; then by (3} and (4) ; and so on. 

Thus, the double-wye interphase circuit reduces the peak current 
to be carried by each tube and by each phase of the transformer wind- 
ing. The utility factor which indicates how effectively the trans- 
former copper is used is increased, as compared with the diametrical, 
6-phase arrangement, Fig. 14. Still, it should be kept in mind that 
the secondary of even a double-wye, interphase transformer requires 



about 40 per cent more copper than would the secondary of a stand- 
ard power transformer of equivalent capacity. 




ARY (b) 


PHASES X| 8 X 2 



i 1 \ 

\ i\ \ 

(O (d) 

FIG. 14. Six-phase (diametrical) rectifier circuit illustrating the theoretical 
and practical wave shapes of d-c current and voltage. 

Fig. -16 shows another very interesting arrangement of a rectifier 
circuit. The transformer has a conventional 3-phase wye-connected 



Vol 45, No. 6 

secondary. With 6 tubes connected as shown we have, in effect, 
two 3-phase rectifiers : one connected to the positive d-c bus ; another 
to the negative d-c bus. The total effect on the d-c system, as far as 
the ripple is concerned, is the same as if we had a conventional 6- 
phase rectifier. 

The transformer winding is, however, better utilized since each 
phase carries current in both directions, and therefore the "utility 
factor" is greater. Hence it is named "2-way circuit." But, while 
the transformer duty is lightened, that of the tubes is increased : each 




FIG. 15. The most popular power rectifier circuit known as "delta-double 
wye" with interphase transformer. 

tube carries full current for l / s cycle. Therefore, larger tubes are 
needed with this arrangement than with that of Fig. 15, everything 
else remaining constant. The arrangement Fig. 16 lends itself quite 
readily to a 3- wire system. The neutral of the transformer (see 
dotted line) can be directly connected to the neutral bus ; with 250 v 
between the positive and the negative buses, we have 125 v between 
each of these buses and the neutral. Each 125-v half of the system 
can be operated independently of the other; i.e., one can be fully 
loaded, with the other carrying no load at all. But, one should re- 
member that each half of the rectifier, as shown on Fig. 8, is only a 
3-phase unit, and therefore the d-c ripple appearing on each- 125-v 
system is greater than that between the 250- v buses. 






We have already mentioned that the ignitron tube must be fired 
every cycle. The igniters with which each tube is equipped require 
for the short time of "firing" about 50 amp peak current at about 
350 v peak voltage. This should be d-c, or unidirectional current, 
with the igniter's potential being positive in respect to the mercury 
pool. Fig. 17 shows one method of accomplishing this, which is by 
far the most widely used in rectifier work; the diagram refers to a 
6-tube equipment, and the power 
circuit is not shown for the sake 
of clarity. 

An auxiliary control trans- 
former (ET) is connected to the 
same source of power as the main 
power transformer; this estab- 
lishes synchronism between the 
power and the control circuits. 
Each phase of the excitation cir- 
cuit excites 2 ignitrons whose 
anode voltages are 180 degrees 
apart in phase relation. 

The secondary of the trans- 
former (ET) energizes 2 net- 
works per phase: (1) a network 
for generating impulses which, 
when fed to the igniters, causes 
them to fire; (2) an adjustable 
phase-shifting network which 
provides a convenient means of 


FIG. 16. "Double- way "rectifier cir- 
cuit, which gives high utilization fac- 
tor of the transformer, and lends itself 
readily to the 3-wire d-c service. 

phase-shifting the firing point and, consequently, a means of adjust- 
ing the d-c voltage output of the rectifier. 

The first of these networks consists of a linear reactor (LL), satu- 
rating reactor (FL), and capacitor (FC). The linear reactor is de- 
signed to give constant reactance up to rated voltage and frequency 
of the circuit. The saturating reactor is designed to saturate when 
the capacitor is charged to the maximum voltage, thus allowing this 
capacitor to discharge through the saturating reactor into the igniter. 
The effect is shown on the insert of Fig. 17; igniter current curve has 
a sharp and pronounced peak which "fires" the tube. As quickly as 
the cathode spot is formed, an auxiliary anode shunts both the dry 



Vol 45, No. 6 

plate rectifier and the igniter, collecting the arc and thereby relieving 
both igniter and the dry plate rectifier of considerable duty. 

The adjustable phase-shifting network consists of a saturable re- 
actor (SL), linear reactor (CL), and capacitor (PC). The reactance 
of the reactor (SL) is adjusted by means of d-c excitation ; the value 




b a 4 

V V 


FIG. 17. Excitation circuit for a 6-tube rectifier. The insert shows 
how the phase retard of firing affects the d-c voltage. 

of this excitation shifts therefore the phase angle of voltage (E) used 
for firing; for instance, from point A to B. This means that the 
tube will conduct current for only the shaded portion of the cycle. 
Obviously, the average voltage is reduced when the firing point was 
moved from A to B. 

Thus, we have a convenient means of adjusting and controlling 
the output d-c voltage of the rectifier within certain limits, indepen- 
dently of the a-c supply voltage. For instance, by providing a volt- 


age regulator GDD (see Fig. 17), the conventional, industrial rectifiers 
keep the d-c voltage constant regardless of wide fluctuations of load 
and with a-c voltage varying about 5 per cent. 

Generally speaking, the phase control of firing can be extended to 
provide 100 per cent voltage control of rectifier, from zero to rated 
voltage. There are, however, several points which should be noted 
in this connection : 

(a) Phase retarding distorts the wave form of d-c output. 

(6) It lowers the power factor on the a-c side in about the same proportion 
as the d-c voltage is reduced. 

(c) The average current carrying capacity of the tubes is somewhat reduced 
owing to the wave distortion; in other words, if a given rectifier can carry con- 
tinuously, say, 1200 amp d-c at full voltage of 250 v (with no phase retard, or with 
a small amount of it), it may be good for carrying something less than 1200 amp 
at very much reduced voltage. 


The primary purpose of d-c supply is to take care of lighting re- 
quirements. Since many of these lights are of high intensity, the 
use of arc lights is indicated. . To avoid any noise or whistle associated 
with a-c arc lights (and this noise would be picked up by the sound 
track), the use of d-c arcs is imperative. Even with d-c used, special 
attention should be given to the ripple which might exist in the volt- 
age curve, since even a small ripple might be recorded as noise. This 
is particularly true in Technicolor practice when the light intensity is 
much higher than average. 

A d-c arc requires about 90-100 v for its maintenance and calls for a 
ballast resistance for arc stability. In order to save on distribution 
copper, it is customary to provide a 1 25-250- v, 3- wire power supply 
and distribution, with arc lights being connected between the line 
and neutral leads. 

Three-unit sets, with two 125-v generators connected in series and 
driven by one synchronous motor, have been and are widely em- 

Large studio establishments have many stages, several of which 
may be in operation simultaneously; it is then customary and logical 
to install several motor generators in a centrally located substation 
supplying d-c power to the stages by means of buses. In this manner 
advantage is taken of the load diversity factor between the several 
stages, and better continuity of service is assured by having one or two 
spare units in this central location. Substations with 6, 8, or even 

432 L. A. UMANSKY Vol 45, No. 6 

10 motor-generator sets, each of capacity up to 500 kw, are encoun- 
tered. Of course, if the area to be covered is too large, and blocks 
of power to be transmitted are appreciable, then more than one sub- 
station can be readily provided. 

In many instances portable motor generators, mounted on rubber- 
tire trailers, are successfully employed. They can be readily moved 
from place to place wherever work is being carried, and this advantage 
is obvious, particularly for temporary locations. 


A d-c generator is, in effect, an a-c machine, whereby the a-c cur- 
rent generated in the armature is mechanically rectified by the com- 
mutator. An oscillograph shows that the d-c voltage thus obtained 
is not a strictly straight line like a storage battery would supply, 
for instance but includes many small ripples. The ripple on com- 
mercial generators of fairly large size is usually kept under 2-3 per 

The predominating frequency of the ripple is produced by the 
armature slots. To minimize the ripple, the designers take recourse 
to the skewing of the slots, or to proportioning of the slot pitch and 
the pole arc. By these and other means special machines can be pro- 
duced for studio work whereby the ripple is reduced to about l /% per 
cent. A further reduction of the ripple is usually obtained by ex- 
ternal filters, located in the generator leads or in the feeders, and at 
times in series with the individual lights. 


Three-Wire Arrangement. If a rectifier is considered for this 3- 
wire service, with as good results as with motor-generator sets, 
careful attention should be given to the number of phases used. 

If it is necessary to operate each 125-v half of the equipment inde- 
pendently, then the ripple should be also taken care of for each hah 7 

Fig. 18 illustrates an arrangement made for a 500-kw, 125-250-v 
unit. The transformer has 2 cores mounted in one tank, each core 
with a separate primary winding, one delta, the other wye. With 
the secondary windings connected as shown, we have, in effect, 12 
tubes in each 125-v circuit. Each of the 4 secondary windings is 
connected to a 6-tube group similar to the arrangement on Fig. 16. 
The positive and the negative terminals of each rectifier group are 

Dec., 1945 



interconnected, as shown, through reactors, acting as interphase 
transformers. The neutral bus is connected to the neutrals of trans- 
former secondaries. 



TRANSF-. c.B. 


5 4 















, r 






] f 

















"C) 1 













L I2 

-, 1 





_ / 


C. B. 


^C I2 


C|2 c . 

::: =; 


L 24 



125 V 

L 24 




FIG. 18. Arrangement of a 500-kw, 125-250-v rectifier for supplying 
power for studio lights. Twelve-phase operation is provided for each 
125-v system. 

Two voltage regulators and 2 excitation circuits are employed, one 
for each 125-v side. Each regulator is acting in the previously de- 
scribed manner on the phase shifting of the 12 tanks, connected either 
to the positive or to the negative bus. With the load varying from 
zero to 100 per cent normal, and with the a-c supply voltage varying 



Vol 45, No. 6 

5 per cent, not more than 15 per cent phase retard is needed to 
maintain a constant d-c voltage, within one per cent of pre-set value. 




FIG. 19. Arrangement of a 2-wire rectifier operating in parallel with 
several 3-wire motor generators. 

It is worth noting that the rectifier, being an electronic device, has 
practically no inertia in responding to the voltage regulator. 
With this layout, i. e., with a 12-phase circuit used for each 125-v 
system, the predominant ripple frequency will be 720 cycles and then 
1440 cycles, if the power supply is 60 cycles. In order to bring the 

Dec., 1945 



magnitude of this ripple to a value obtainable with good motor gen- 
erators, the rectifier will be equipped with a tuned filter, or a resonant 
shunt, connected as shown to each 125-v system. The reactances 
LIZ and L 2 4, and the capacitors C\z and Cu are selected to "drain off" 
the 12th and the 24th harmonics in the d-c circuit. The reactors LI 
act as current limitors for this high-frequency current. 


FIG. 20. Rectifier substation with phase multiplication 
increased by primary connection of transformers. 

Voltage Adjustment. This rectifier can be readily operated at re- 
duced voltage, say down to 50 per cent normal, or still lower if needed. 
This can be provided by means of phase control, at some sacrifice of 
d-c form wave. Even then a well-designed tuned filter is capable of 
keeping the ripple within the acceptable limits. 

The following remarks are in order. Arc lights, by their nature, 
are not operated at voltage, adjustable within wide limits. If wide 
control of voltage is indicated, as, for instance, for fade-outs or other 
lighting effects, then incandescent lamps are employed. In this case 
the problem of d-c ripple affecting the sound track does not exist. 

Thus, it seems logical to concentrate on limitation of ripple exist- 
ing at about normal voltage on the system. 



Vol 45, No. 6 

Two- Wire Arrangement. The 3-wire rectifier just described is 
self-contained; i. e., it can give complete d-c service independently 
of other sources of d-c power; it has a sufficient number of phases 
to keep the ripple under control. 

We can consider another case when a rectifier unit is added to an 
existing 3-wire d-c system of several times the capacity of the new 
unit. Let us assume that the power supply is now provided by 
several 125-250-v motor generators (see Fig. 19). If it is feasible 
to always operate several of these sets whenever the new rectifier is 
connected to the same bus, then it might not be necessary to make the 





FIG. 21. Power factor of a rectifier ; note that it remains 
high even at light loads. 

rectifier of the 3-wire type; a simpler, 250-v, 2-wire unit might be 
satisfactory. In this case a 12-tube, 12-phase assembly is quite 
feasible not greatly different from a conventional industrial rectifier 
of the same capacity. 

Phase Multiplication by Rectifier Groups. Consider now a brand 
new installation, say a substation with eight 300-kw, 125-250-v 
rectifiers, for studio lighting and no motor-generator sets in the 
same station. It is feasible then to provide the necessary "phase 
multiplication" for the whole substation rather than for each recti- 
fier. This may offer some advantages in certain specific cases. 

For instance, a 300-kw, 125-250-v rating can be readily provided, 
see Fig. 20, by twelve 200-amp tubes arranged as shown. This gives 



us only a 6-phase operation either between any one bus and the 
neutral, or between the 2 outside buses. 

By arranging the second 300-kw unit in the like fashion, but with 
its transformer primary connected wye instead of delta, we get a 12- 
phase operation when both rectifiers, No. 1 and No. 2, are operating 
together. The remaining rectifiers can be arranged likewise and 
operated in pairs. 

If the units involved were of 500-kw, 125-250-v rating, then each 
would require twenty-four 200-amp tubes for the sake of current 
carrying capacity, and a 12-phase operation for each 125-v circuit 
would be available without extra tubes (see Fig. 18). However, even 

FIG. 22. Portable rectifier substation, rated 500 kw, 125-250 v, used at Navy 
docks for servicing ships. 

then the arrangement in Fig. 20 has the advantage of simpler trans- 
formers. On the other hand, it restricts the freedom of the operator 
in selecting at random the rectifiers which he may wish to connect to 
the bus at any given time; he should use them in pairs to get a 12 
phase performance. 

Power Factor. The power factor of a rectifier is high, but is lag- 
ging. It does not fall off rapidly with diminishing load, like, for in- 
stance, is the case of an induction motor. It usually is between 90 
and 95 per cent, and stays that high down to 25 per cent load or even 
lower. Fig. 21 illustrates this characteristic of the rectifier. 

Parallel Operation. There is no difficulty in operating a rectifier 
in parallel either with other rectifiers or with motor-generator sets. 
Experience in many industries has fully demonstrated that such 
operation is quite successful. Voltage and load regulators acting on 



Vol 45, No. 6 

the excitation circuit of the rectifiers take care of this performance. 
Mechanical Arrangement. For units of such capacity as are, 
and as probably will be, considered for this application, the sealed 
tube rectifiers are indicated. For the sake of phase multiplication 
it is to our advantage to use, within economic limits, a larger number 
of smaller tubes rather than a smaller number of larger units. As 
we have already seen, a 500-kw, 1 25-250- v rectifier designed for 
12-phase operation in each 125-v leg would be built with twenty- 
four 200-amp tubes. 



I 70 
| 60 
<n 50 





3 V 


















FIG. 23. Effect of d-c voltage on the efficiency of the rectifier. 

It is best to build such equipment along the lines which have found 
full acceptance for other industrial applications of rectifiers. The 
several tubes are mounted in steel cabinets, with the firing circuits 
and accessories located in the rear of the same cabinets. The power 
transformer will be mounted adjacent and will be "throat connected" 
to the rectifier compartment. The d-c switchgear, voltage regulators, 
and filters will be located in adjacent cubicles. 

This "metal clad" construction consists, therefore, of factory- 
built units which are completely wired and can be erected with mini- 
mum of time at the destination. Since there are no rotating parts 
involved, no foundation is required. 


Fig. 22 illustrates a 500-kw, 1 25-250- v portable rectifier, one of 
several units built for Navy docks, to supply d-c power to berthed 
ships. It shows that similar construction is quite feasible for studio 
work if a portable unit is called for. 


In this review we have shown that the engineering problems in- 
volved in applying rectifiers to the studio d-c power systems can be 
readily solved. Rectifiers with their associated equipment give a 
reliable source of power, easily controllable and free of voltage ripple. 
But so do the motor-generator sets. What, then, should be the basis 
for the choice between the 2 types of converting equipment? 

It should be admitted to begin with that in this competition with 
the motor-generator set the rectifier starts with a heavy handicap. 
After all, we are calling for a low-voltage d-c supply, such as 125 v. 
Thus, for a given kilowatt rating the ampere rating is higher than 
would be, say, with 250 v, not to speak of still higher voltages. Since 
the electronic tubes are rated on ampere basis, the low operating volt- 
age calls for a larger number and a larger capacity of tubes. This, 
obviously, affects the cost. 

The 125-v rectifiers are not as efficient as those built for higher 
voltages, primarily on account of greater effect of the arc drop. Fig. 
23 illustrates this fact quite well. However, even then the part- 
load efficiency is better than with motor-generator sets, and this fact 
is important in case the load is intermittent. 

The low light load losses of the rectifier as compared with rotating 
equipment may tend to justify the selection of larger units. For 
instance, instead of having, say, six 300-kw motor-generator sets, it 
is advisable to provide only four 500-kw rectifiers ; it is less important 
to reduce the capacity of units connected to the bus at times of light 

The rectifier equipments are easier to install than the rotating 
equipment. No foundations are required. Motor generators, or 
the substation where they are installed, should be ventilated, and 
suitable provisions should be made. On the other hand, the rectifiers 
are water-cooled. In many cases this arrangement is more satis- 

The rectifier substation need not be attended continuously since 
the entire equipment is stationary. This condition does not neces- 
sarily hold true for motor-generator sets. 

440 L. A. UMANSKY 

In case of temporary a-c power failure, the service may be restored 
quicker with the rectifiers, by simply reclosing the primary breaker. 
The several motor-generator sets must be restarted and resynchro- 
nized, usually one at a time, and this causes delay. In many industrial 
plants this feature is considered as a definite advantage of the rec- 

Thus, the motion picture industry can readily accept the rectifiers 
as fully suitable for this application. Of course, each specific case 
should be treated on its own merits to make certain that the best type 
of equipment has been chosen. 



Summary. A small microphone boom having the versatility and operating 
controls of all location and other booms is described. The rear end of conventional 
booms extends substantially beyond the outside of their dollies and the pivot mast 
protrudes well above the top of the boom pole limiting the overhead clearance. The 
small boom solves this condition by being so designed and proportioned that with a 
minimum of overhead clearance and rear end overhang, and together with its special 
perambulator, it can be used on smaller sets where it is necessary to place the boom 
in a corner, or to play the transmitter up close to a low ceiling to keep it out of the 

The perambulator frame telescopes to keep within the limits of the boom and to 
permit easy handling and transportation. 

Motion picture settings have always strived for the greatest detail 
and realism. Owing to wartime restrictions of material, many of the 
sets are being constructed smaller than it is convenient to photograph 
and record in. This has required the development of a small micro- 
phone boom which can be operated in such small sets without any 
penalties or handicaps whatsoever in its performance. 

The microphone boom described in this paper has been designed 
to meet the following specifications for use in small sets : 

(1) It must be adjustable in height so that it may be dollied through a 7-ft 
door, or be raised so that the operator can see over the lamps and cameras into the 

(2) Its width must be such that it will pass through a 30-in. door, but it must 
be extremely stable at all times. 

(5) The boom and its perambulator must be capable of being operated within 
30 in. of a wall which is the minimum distance in which a microphone boom 
operator can stand and work. 

(4) It must be capable of good pickup as close as 6 ft from the operator, or as 
far away as 12 ft. 

(5) It must operate as quietly and with every facility provided by the larger 
booms in common use. 

* Presented May 18, 1945, at the Technical Conference in Hollywood. 
** Warner Bros. Pictures, Inc., Burbank, Calif. 


442 B. F. RYAN AND E. H. SMITH Vol 45, No. 6 

In Fig. 1, the boom proper consists of a fixed outer steel tube to 
which all the operating controls are attached. It is mounted in a cast 
yoke on a horizontal trunnion bearing which permits an up or down 
movement through an angle of 50 degrees in addition to a horizontal 
rotation on a vertical bearing in the yoke. 

The sponge rubber padded arm rest, as seen in Fig. 2, is held lightly 
under the operator's left arm by presetting the balance so that the 
front end is slightly heavy at all times. The cable drum, operated by 
a hand crank, is of such proportions that 3 complete turns of the 



FIG. 1. Side view of small microphone boom with operator in position. 

handle will rack the inner boom pole from the closed to the fully ex- 
tended position, a travel of 6 ft. 

The left-hand controls the "twist-gag" handle which, through a 
single pull cord, rotates the microphone hanger drum, Fig. 3, up to 
one and one-half turns. The unidirectional microphone, which is 
customarily used, will then always be in the proper position for the 
best quality of pickup. 

In a further effort to reduce the distance between the top of the pole 
and the bottom of the microphone, the hanger, Fig. 4, and twist-gag 
arrangement were redesigned. Fig. 5 shows a partial cutaway sec- 
tion of the hanger and method of control. The microphone is held 
in a ring clamp that is suspended on rubber shock cord and laced to 

Dec., 1945 



an outer ring, which in turn is pivoted on a t/-shaped bracket. This 
hanger assembly can easily be taken off the drum to change to a 
"rain hat" hanger by removing 
a knurled knob. Wind screens 
are mounted directly on the mi- 

The short "pig-tail" cable is 
ordinarily held in place in the 
large hole through the drum 
bearing by a sponge rubber bush- 
ing; however, the cable, being 
under considerable stress from 
the oscillating movement of the 
microphone hanger, may develop 
an open circuit in the shield, in 
such case it may be readily re- 
placed. No tools are required, 
as it has a Cannon plug at one 
end and a receptacle with a 
gland nut at the other that may 
be loosened and the receptacle 
and cable drawn through the hole. 

One end of the cord used to 
rotate the assembly is fastened 
on a fixed sheave bracket then 
passed over sheaves and around 
the hanger drum, then down the 
inside of the boom pole, and 
finally spliced to a length of rub- 
ber shock cord anchored in the 
pole. The composition of the 
shock cord being basically rub- 
ber, its life is considerably pro- 
longed by placing it within the FIG. 2. Rear view of small microphone 


tube away from light and mois- 
ture. This arrangement replaces the former method of a coiled 
clock spring which had to be covered with a graphite impregnated 
cotton sleeving and mounted inside the hanger. Though reasonably 
satisfactory, it was, nevertheless, a source of noise and frequent 



Vol 45, No. 6 


All moving parts on the boom are mounted in babbitt bearings 

_, and lubricated by Zerk fittings. 

This type of bearing material 
has been found by numerous ex- 
periments to give the quietest 
and most trouble-free service. 

The type and speed of all 
operating controls are identical 
with those on the larger booms ; 
thus, the boom man can switch 
back and forth between the large 
and small booms without notice- 
able difference in the feel or 
manipulation of the controls. 

Fig. 6 shows the direct com- 
parison between the Warner Bros, 
standard size and small booms, 
both in the closed position. As 
a standard of comparison, the 
microphone holder on the small 
boom is 6 ft from the center post. 
The small boom has an over-all 
closed length of 8 ft 6 in. in- 
cluding the rear-end overhang 
which, at all times, remains 
within a radius of 30 in. from the 
center column. 

The frame of the small peram- 
bulator is of welded steel tubing 
construction. Though shown in 
the mid position, the wheelbase 
is adjustable from an over-all ex- 
tended length of 4*/2 ft to a 
closed position of 3 ft, the steer- 
ing arm nesting in the recess 
in the platform to keep within 
the rear-end overhang of the 

In Fig. 7 both booms are 
The maximum reach of the small boom from 

FIG. 3. 

Front view of microphone 

fully extended. 

Dec., 1945 



the center post is 12 ft, which is 6 ft less than that of the large 

FIG. 4. Microphone hanger. 

FIG. 5. Detail of microphone hanger. 

Fig. 8 shows the final assembly drawing containing the principal 
dimensions and their limits. A fabric covered flexible steel cable, 



Vol 45, No. 6 

or standard tiller rope, is wound on an aluminum drum, then reeved 
over a series of sheaves and secured to the rear end of the inner, pole 
to move it in and out. The cable is also fastened to the counterweight 

FIG. 6. Comparison of large and small microphone booms; shafts retracted. 

FIG. 7. Comparison of large and small microphone booms; shafts extended. 

carriage through a pair of threaded thimbles providing for any neces- 
sary adjustment. The trunnion bearing is set 6 in. forward to allow 
the use of a smaller fixed counterweight. This stationary weight is a 
55-lb cast block of lead attached to the rear end of the outer tube and 

Dec., 1945 



is made removable leaving a net weight of 82 Ib for the boom, thus 
facilitating its handling and transportation. 

The traveling counterweight on the boom serves a dual purpose 
in that it controls the slack in the microphone cable and twist-gag 
cord, as well as compensates for the out-of-balance condition that 
develops during the extension of the boom. As it was impractical to 
mount the traveling weight in the rear and meet the requirement that 
the boom could operate 30 in. from a wall, the traveling weight was 
placed in front of the vertical shaft. 

The carriage rides through rubber guide rollers on 2 small steel 
tubes suspended from the underside of the large outer tube. A two- 

FIG. 8. Dimensional drawing of small microphone boom. 

to-one ratio was necessary to move the carriage, and it was obtained 
by placing another grooved drum in the center of the trunnion fork 
and on the same shaft with the main cable drum, but having exactly 
one half its pitch diameter. The original arrangement, with the 
carriage forward and the boom fully in, is balanced by the fixed weight 
on the rear end. The weight of the carriage, however, had to be such 
that its moment would decrease by the same amount as the moment 
of the inner pole increased when it was boomed out and vice versa. 

In order to obtain the most advantages from this type of boom it 
was necessary to design a special perambulator that combined sta- 
bility with the ability to be collapsed to smaller proportions and 
thus remain within the limits of the boom. In this condition it would 


be possible for an operator to ride on the platform with the drop 
leaf folded back and control the boom while it is being dollied through 
a 30-in. door which is 6 ft 8 in. high. By means of the crank on the 
center column operating through a self-locking worm and rack and 
pinion, the boom can be raised to a maximum height of 7 ft 9 x /2 in. 
above the floor. Steering is accomplished through a forked tee handle 
mounted on the rear caster wheel. The parking brake consists of a 
shoe operated by a cam mechanism and is applied to the tire auto- 
matically when the steering handle is latched in the vertical position, 
but may be readily released by a downward pressure on the catch. 
Primarily the small-type boom was designed for use where space 
limitations brought about a condition whereby it was impractical to 
use the standard boom. However, this does not preclude a broad 
useful field for this boom, as emphasis was placed upon standardiza- 
tion of most parts and accessories in order that they could be freely 
interchanged, any boom fitting any perambulator, and, if necessary, 
they will fit the standard lamp tripod. 



Summary. A crater brightness of as much as 1400 candles per sq mm is pro- 
duced by a new 13.6-mm super high-intensity positive carbon. This brightness is 
obtained at, 290 amp using water-cooled jaws of special design. With conventional 
positive carbon jaws, a crater brightness of 1200 candles per sq mm is obtained at 265 
amp. The burning rate of the positive carbon in both cases is approximately 45 in. 
per hr. 

Tests of these carbons with a relay condenser optical system indicate possibility of a 
30 to 45 per cent increase in quantity of screen light for transparency process projection. 
Measurements with a standard condenser optical system of the type used in theater 
projection show increases of 40 to 60 per cent over standard carbons. Mention is 
made of the necessity of adequate provisions in order that the components of the pro- 
jection system can accommodate the faster burning rate, higher power and higher 
intensity of radiant energy associated with the operation of the new carbons. 

Technological developments of the past 10 years have resulted in a 
succession of advances which have increased several fold the quantity 
of light which can be projected on a motion picture screen. This is 
true both for transparency process projection 1 - 2 in the motion pic- 
ture studios, and also for projection in motion picture theaters. 
These advances have resulted from improvements of the various com- 
ponents of the projector system including the carbon arc light 
sources, 3 ' 4t 5 arc lamps, optical systems, projectors, etc. Present 
theater projection systems make it possible to project 2 to 3 times as 
much light to the screen as was possible 10 years ago. Light levels 
with modern studio transparency process projection equipment can 
be as much as tenfold those of a few years ago. However, the 
industry has used all the light that is available and has expressed a de- 
sire for more. This paper will demonstrate that still further increases 
in screen light can be obtained owing to the recent development of a 
new 13.6-mm experimental positive carbon which operates with a 
higher crater brightness than any carbon commercially available at 

* Presented May 14, 1945, at the Technical Conference in Hollywood. 
** National Carbon Company, Inc., Fostoria, Ohio. 



the present time. This new 13.6-mm super high-intensity positive 
carbon is a result of intensive research and development work per- 
formed during the past few years and directed toward the goal of 
higher brightness. 

Operating Characteristics. This paper is concerned with the 
type of high-intensity arc and carbons wherein the positive carbon 
is rotated during burning and the negative carbon is placed at an 
angle with respect to the positive. At the present time standard 

FIG. 1. Special water-cooled jaws for positive carbon 
showing S, silver contact blocks; /, water jacket; P, posi- 
tive carbon; and N, negative carbon. 

National 13.6-mm super high-intensity projector positives 3 and 
National 16-mm super high-intensity studio positive carbons 5 are 
employed to provide the highest levels of screen illumination for 
background projection, and the 13.6-mm carbons to give the maxi- 
mum screen light for theater projection. These 13.6-mm and 16-mm 
carbons operate at 170 amp and 225 amp, respectively. In compari- 
son, the new 13.6-mm super carbons have been burned at currents up 
to approximately 290 amp. 

Research has shown that the brightness of the high-intensity car- 
bon arc depends, among other things, upon the density of current 


entering the crater. Increased current density signifies greater con- 
centration of electrons, positive ions and excited atoms in the crater 
gases which are the principal source of light. Following this prin- 
ciple, carbon compositions and methods of burning have been de- 
veloped which allow a much greater dissipation of energy per unit 
area within the crater and results in increased brightness. The new 
13.6-mm positive carbon has been designed in this fashion. It has 
been operated at currents up to approximately 265 amp in lamps with 
conventional air-cooled positive jaws. 

FIG. 2. Close-up of silver contact blocks and mounting studs. 

However, the composition of the new 13.6-mm super carbon is such 
that its maximum current rating can be extended to currents higher 
than 265 amp by employing methods which will more effectively cool 
the positive carbon. One means of accomplishing this result has 
been the use of water-cooled jaws of a special design in combination 
with a short protrusion of the end of the positive carbon beyond the 
jaws. The important features of this design are illustrated in Fig. 1. 
The cooling water in the jackets, /, comes into direct contact with 
the silver blocks, S, which fit snugly around the positive carbon, P. 
Fig. 2 shows in more detail the design and construction of the silver 
contact blocks. This design coupled with the high conductivity of 
silver permits unusually rapid removal of heat from the carbon. 



Vol 45, No. 6 

Flow rates of water of approximately one gallon per min are more than 
ample to take care of any of the operations described in this paper. 

A new 5 /8-m. copper-coated negative has been developed to operate 
with the new 13.6-mm carbon. This negative carbon can be used in 
the conventional manner. Another type of negative also has been de- 

FIG. 3. 

5432 234567 


Distribution of brightness across crater of 13.6-mm super high intensity 

veloped for use with the new positive carbon. This is an unplated 
7-mm negative carbon which is designed to operate in special water- 
cooled jaws. This type of negative carbon has ample current- 
carrying capacity for operation over the 265-290-amp range. The 
absence of a copper plate eliminates any possibility of copper drip- 
pings adhering to the lamp optical system. The cooling water dissi- 
pates heat which otherwise would be absorbed by lamp parts. Im- 
proved arc stability is obtained through use of the smaller diameter 
negative carbon, 

Dec., 1945 




Burning Rate 

of Positive 
Protrusion of Carbon Crater 
ositive Carbon Amp Volts In. Per Hr Cp 





T t 






i i 


> 1 




















H 1 



































ive Carbon 







Copper Coated 

Copper Coated 









































itive Carl 



I 1 











2 >> 












I i W 








g ^ 




Studies have been made in a laboratory test lamp on the burning 
performance of the new 13.6-mm super carbons in comparison with 
the standard 170-amp 13.6-mm and 225-amp 16-mm super carbons. 
Crater brightness and total crater candlepower were measured using 
the method and equipment recently described. 6 The 3 types of car- 
bons were operated both with air-cooled and with the special water- 
cooled positive carbon jaws. The results are given in Table 1 and 
Fig. 3. 

Fig. 3 shows the brightness distribution across the crater for the 
13.6-mm carbons. The new 13.6-mm carbon at 265 to 290 amp has a 
brightness at the center of the crater ranging approximately from 
1200 to 1400 candles per sq mm. These values, respectively, are 
about 30 and 50 per cent greater than the brightness of the 170-amp 
13.6-mm carbon. A maximum operating current of 265 amp was 
obtained for the new 13.6-mm carbons with a ! 3 /8-m. protrusion 
both with air-cooled jaws and with special water-cooled jaws de- 
scribed above; the burning characteristics were practically identical 
with both types of jaws. However, a reduction in protrusion to */2 
in. with the special water-cooled jaws allowed the maximum operating 
current to be increased to 290 amp. 

Some insight into the significance of the combination of water- 
cooled jaws and short protrusion may be obtained from measurements 
of the amount of heat carried away by the cooling water. With a 
positive carbon protrusion of 1 / 2 in. and a current of 290 amp, the 
amount of power carried away as heat by the cooling water was 4.2 kw 
which is approximately 18 per cent of the input power to the arc. 
When the protrusion was increased to l 3 /s in. with a current of 265 
amp, the power carried away as heat decreased to 2.8 kw which is 13 
per cent of the arc power. This difference in heat transfer by the cool- 
ing water made it possible to burn the new carbon at the higher 

The data in Table 1 show that the burning rate of the new 13.6-mm 
carbon is 45 in. per hr, or approximately double that of the standard 
13.6- and 16-mm carbons. It is significant to note that there was no 
increase in burning rate of the new 13.6-mm carbon with the increase 
in current from 265 to 290 amp. The explanation for this observa- 
tion undoubtedly rests on the improved cooling and reduced oxida- 
tion afforded by the decrease in protrusion which accompanied the 
change in current. 


Application of New Carbons to Transparency Process Projec- 
tion. There is general recognition of the important role of trans- 
parency process projection in modern motion picture production. 
Although improvements in recent years have greatly expanded the 
usable area of process projection screens, the present possibilities 
are often less than desired. It was visualized that the new 13.6-mm 
super carbons might offer a significant improvement in quantity of 
screen light available for background projection. Consequently, 
arrangements were made to obtain information on this point. 

Through the cooperation of Farciot Edouart of Paramount Studios, 
Hollywood, we were able to test the new 13.6-mm carbons in the 
Paramount transparency process projection equipment. This equip- 
ment contains the Paramount design of relay condenser system which 
has shown such merit in process projection. This equipment also 
employs the Mole-Richardson arc lamp designed to Academy Research 
Council Process Projection Specifications. This lamp utilizes water- 
cooled positive carbon jaws and head and the positive carbon protru- 
sion from the jaw was ! 5 /s in. Under these conditions it was found 
that the new 13.6-mm super positive carbon and 5 / 8 -in. negative 
carbon could be burned at currents as great as 265 amp. As bases of 
comparison, tests were also made with the commonly used 16-mm 
super high-intensity studio positive at 225 amp, and with the 13.6-mm 
super high-intensity projector at 170 amp. The steadiness of the 
light was practically equivalent on all carbons at the indicated cur- 
rents. The results of the measurements are given in Table 2 and 
show that the new carbon offers approximately a 30 per cent increase 
in screen light over the 16-mm standard carbon at 225 amp, and a 45 
per cent gain over the 13.6-mm 170-amp carbon. Distribution of in- 
tensity on the screen was good and was quite comparable with all the 
combinations employed. 

Screen Light for Transparency Process Projection 

Positive Carbon Amp Relative Screen Lumens* 

Standard 13.6-mm Super High-Intensity Projector 170 90 

Standard 16-mm Super High-Intensity Studio 225 100 

New 13.6-mm Super High-Intensity 265 130 

* Measured at Paramount Studios with relay condenser system and silent 
camera aperture (0.723 in. X 0.980 in.). 

This new carbon, which offers increases in illumination of from 30 to 
45 per cent over that obtainable from present standard carbons, 



Vol 45, No. 6 

should make available a significant increase in usable screen area. 
It is estimated that an additional 10 to 15 per cent increase in screen 
light would be expected from operation at 290 amp which can be 
made possible as described above. 

Application of New Carbons to Motion Picture Theater Projec- 
tion. The new 13.6-mm positive carbon was also considered with 
respect to the quantity of screen light available for 35-mm film 
projection with a standard condenser optical system such as used 
in theaters. The 13.6-mm super high-intensity projector carbon 
and the new 13.6-mm carbon were compared in laboratory tests 
using a standard 35-mm film aperture and//2.2 condensers (operated 
at//2.0 distances) and with a treated 5-in. focal length //2.0 projec- 
tion lens. As shown in Table 3, the measured screen light without 
shutter or film was increased from 18,500 lumens for the 13.6-mm 170- 
amp carbons to 26,000 lumens with the new 13.6-mm positives oper- 
ated at 265 amp and to 30,000 lumens at 290 amp. This should 
make available 40 to 60 per cent more screen light than the maximum 
now obtainable. 


Screen Light for Motion Picture Projection 


Standard 13.6-mm 
Super High-In- 
tensity Projector 

New 13.6-mm Super 

New 13.6-mm Super 

Carbon Jaws 

tooled or Special 

of Positive 






Intensity of 
Energy at 
Center of 
Film Aper- 
ture Watts 
Per Sq Mm** 



Air-cooled or Special 

Special Water-cooled 


265 26,000 

290 30,000 


* At 80 per cent side to center distribution without shutter, film or filters, and 
with standard 35-mm (0.600 in. X 0.825 in.) aperture; f/2.2 condensers and// 
2.0 treated projection lens. 

** Radiant energy measurement made with system adjusted to give maximum 
intensity at center of film aperture. 

Determinations of the intensity of radiant energy incident at the 
center of the film aperture were made with the 13.6-mm super and the 
new 13.6-mm carbons in the above condenser optical system. The 
technique used was the same as described in a recent paper 7 and in- 


volves the use of a pinhole aperture, thermopile, and filters. The 
intensity of radiant energy is listed in Table 3 in combination with 
the screen light data obtained. With the new 13.6-mm positive the 
maximum intensity at the center of the film aperture is 1.45 to 1.65 
w per sq mm compared with a value of 1.05 w per sq mm for the 
standard 13.6-mm super high-intensity projector positive carbon 
system. By using a heat filter 7 it is possible to reduce markedly the 
total energy flux at the film aperture with a smaller reduction in light 
intensity. There are filters which reduce the total radiant energy 
approximately 50 per cent with a light reduction of perhaps 20 per 
cent. With such a filter, the total radiant energy incident at the cen- 
ter of the film aperture can be reduced to nearly the same level as for 
125-amp standard 13.6-mm condenser systems with approximately 
twice as much light. 

With a heat filter of 80 per cent light transmission, a 90-degree film 
shutter and 75 per cent screen reflectivity, the 290-amp operation 
should yield a brightness of 12-ft-L, the approximate average of the 
ASA Standard range, at the center of a screen 35 ft in width which is 
25 per cent wider than can be illuminated to this intensity with the 
170-amp 13.6-mm carbon. 

It must be recognized that adequate provisions must be made so 
that all the components of the projection system will accommodate 
the special features of the new carbons. For example, the lamp must 
be adapted to the faster burning rate of the new carbons. The lamp 
parts, the condenser lenses, and the projector parts near the aperture 
must be able to withstand the increased heat from the arc. Heat 
filters or other provisions will be necessary to prevent undesirable 
effects caused by heat on the film. When these factors are properly 
taken into account, these new carbons will offer noteworthy increases 
in light for projection. 


1 EDOUART, F.: "The Paramount Transparency Process Projection Equip- 
ment," /. Soc. Mot. Pict. Eng., XL, 6 (June, 1943), p. 368. 

2 EDOUART, F.: "High-Efficiency Stereopticon Projector for Color Background 
Shots," /. Soc. Mot. Pict. Eng., XLIII, 2 (Aug., 1944), p. 97. 

3 JONES, M. T., LOZIER, W. W., AND JOY, D. B.: "New 13.6-Mm Carbons for 
Increased Screen Light," /. Soc. Mot. Pict. Eng., XXXVIII, 3 (Mar., 1942), p. 229. 

4 LOZIER, W. W., CRANCH, G. E., AND JOY, D. B.: "Recent Developments in 
8-Mm Copper-Coated High-Intensity Positive Carbons," J. Soc. Mot. Pict. Eng., 
XXXVI, 2 (Feb., 1941), p. 198. 

6 JOY, D. B., LOZIER, W. W., AND NULL, M. R.: "Carbons for Transparency 


Process Projection in Motion Picture Studios," /. Soc. Mot. Pict. Eng., XXXIII, 
4 (Oct., 1939), p. 353. 

6 JONES, M. T., ZAVESKY, R. J., AND LOZIER, W. W.: "Methods for Measure- 
ment of Brightness of Carbon Arc," J. Soc. Mot. Pict. Eng., 45, 1 (July, 1945), p. 

7 ZAVESKY, R. J., NULL, M. R., AND LOZIER, W. W. : "Study of Radiant Energy 
at Motion Picture Film Aperture," /. Soc. Mot. Pict. Eng., 45, 2 (Aug., 1945), p. 




Summary. A new developer formula is described which has the property of re- 
sisting the effect of bromide while maintaining good characteristics suitable for picture 
negatives. This developer may be replenished in the same manner as ordinary 
positive developers. By use of this formula the H and D characteristic is held essen- 
tially constant, and control is maintained over the effective emulsion speed of the 
developed film . As much as 100 per cent increase in emulsion speed may be obtained 
while maintaining gamma and grain structure comparable to normal commercial 
negative developers. 

In 1940 at the spring meeting of the Society of Motion Picture 
Engineers, in Atlantic City the author had the opportunity to present 
a paper on the subject of "Mathematical Expression for Developer 
Behavior." 1 In that discussion the object was to derive some indi- 
cation as to the optimum chemical structure for a developing agent 
which is required to have given characteristics suitable for picture 
negative development. At a somewhat later date, the application 
of these theoretical considerations resulted in the selection of a mole- 
cule which should have approximately the desired characteristics. 
By a strange coincidence, when the material was prepared and com- 
bined in a developer formula it had almost exactly the characteris- 
tics which were suggested by the theory. 

Unfortunately, as is usually the case, there were certain undesirable 
features to the developer thus produced and it required extensive 
study to bring forth a commercially usable formula. The various 
unusual characteristics of this formula are to be the subject of the 
present paper. The formula will be designated by the coined term 
"Reidol Formula." 

In present laboratory practice it is desirable that a picture negative 
developer produce a gamma of 0.6 to 0.7 while maintaining a high 

* Presented May 17, 1945, at the Technical Conference in Hollywood. 
** Hollywood, Calif. 




Vol 45, No. 6 

level of emulsion speed and a relatively fine-grain structure. One 
serious difficulty with practically all negative developers is their 
sensitivity to bromide. Bromide ions are released into solution by 
the chemical reaction of development. The photosensitive silver 
bromide is reduced to metallic silver and the bromide ions remain in 
the solution. 

The presence of bromide in the solution acts to depress the inter- 
section point of the H and D family. This results in loss of toe den- 
sities or shadow detail and in a change of the contrast of the developed 

2.0 T.O 


FIG. 1. 

It would appear that a highly desirable state of affairs would exist 
if a developer formula could be found which was relatively insensitive 
to the presence of bromide. In Fig. 1 there is shown an H and D 
family for a representative commercial negative developer without 
bromide. Addition of bromide to this developer would seriously 
hamper its operation. Fig. 2 shows the result of adding 3 grams per 
liter of KBr to this solution. 

Fig. 3 shows a similar H and D family for the Reidol formula in 
which is contained 3 grams per liter of KBr. The intersection point 
of the H and D family is actually higher than that of the normal com- 
mercial developer without bromide of Fig. 1. 

Dec., 1945 



A number of interesting results are obtained with the Reidol for- 
mula, or certain modifications of the formula. The stability with 
respect to bromide results in increased emulsion speed for a given 
gamma as compared with commercial developers. This is partly 
because of the fact that it is possible to work the Reidol formula at 
its highest efficiency at all times. There is in addition the fact that 
the Reidol formula is inherently of a lower gamma-infinity than most 
commercially used developers. 






2.0^ iO 

LOG E - 2& SENS/TOME 7f{ 

FIG. 2. 

With a concentration of the order of 3 grams per liter of KBr, it is 
apparent that the small increments of bromide which are generated 
during the development of film will exert relatively little effect on the 
action of the developer. This characteristic may be advantageous 
in the elimination of the so-called "bromide drag" which sometimes 
causes objectionable streaks in negatives. 

By maintaining a balance of bromide concentration in the solution 
it should be possible to develop several thousand feet of film in each 
gallon of developer solution, provided, of course, a proper concentra- 
tion of developing agent is present to prevent chemical exhaustion. 
There seems to be every possibility of establishing a workable com- 



Vol 45, No. 6 

mercial procedure wherein a simple replenisher is added only to the 
extent of carry-over. As long as the bromide content is kept near 
an equilibrium point the developer will continue to act at its maximum 

Fig. 4 shows the behavior of the Reidol formula with respect to ex- 
haustion. In this test fully exposed film was developed to a density 
of about 2.4. Since the average negative density (over the whole 
film) is approximately 0.6 it means that every 10 ft of fully exposed 

f D 





Z 10 


FIG. 3. 

film is equivalent to 40 ft of normal negative film. The assumption 
was made that the developer is to be consumed at the rate of one 
gallon per thousand feet of film. Accordingly, replenisher solution 
without bromide was added to replace the used developer at the rate 
of about 100 ec for each 10 ft of fully exposed film. It should be 
noted that following an initial loss of speed, the developer character- 
istic levels off to a practically constant value. 

This exhaustion test is only indicative of the possible conditions to 
be encountered in commercial operations. Somewhat different be- 
havior is to be expected in practice since other factors will enter, such 
as aerial oxidation and variations owing to rate of replenishment. 

Dec., 1945 



The Reidol formulas operate at pH values of 7.5 or less. The de- 
veloper will still function at />H values as low as 5.5. Variations of 
the formula may be compounded by adding buffers or weak acids, or 
by altering relative concentrations, and a wide range of characteris- 
tics may be obtained by the choice of different operating conditions. 

In Fig. 5, H and D curves are shown for several conditions of time, 
temperature, and modifications of the formula. It can be seen, 
therefore, that desired characteristics may be obtained under almost 
any conditions of laboratory practice. 


FIG. 4. 

One suggestion which was made at the outset is that the additional 
emulsion speed which it is possible to derive from the Reidol formulas 
would be of value to the studios in simplifying lighting problems. 
This may well be, but it is only fair to state that the additional speed 
may be detrimental in certain cases. For example, an outdoor 
scene may be so bright that the lens must be stopped down to a point 
where the image suffers. The very next scene may be one where the 
maximum film speed is none too great. For practical considerations, 
this wide range of requirements must be handled by one film. 

By use of certain of the Reidol formulas, it is possible for the 



Vol 45, No. 6 

cameraman to choose the emulsion speed at which he wishes to oper- 
ate. By subsequently employing the appropriate development pro- 
cedure it is possible and even quite practical to develop the films to 
the desired characteristic. This may all be done in the same de- 
veloper solution and at the same temperature. The cameraman, 
therefore, has control over the speed of the film he is using and, ef- 
fectively, he is given a variety of different films in the same can. 
The control is effective over a range of more than two to one in speed. 









\ J-*r -65 

2.0 To 0.0 


FIG. 5. 

Fig. 6 shows the possible variations in speed as obtained with the 
Reidol formula. 

The grain structure of the developed image is comparable with 
that obtained from other commercially used picture negative de- 
velopers. The images have a slightly brownish cast which is char- 
acteristic of fine-grain developers. There is more fog developed by 
the Reidol formula than normal since the energy is quite high. How- 
ever, this so-called "veil" is constant and does not impair the image 
quality. The important factor in film speed is, of course, the dif- 
ference between the base or fog density and the actual image density. 
If the ultimate of emulsion speed must be obtained, then a certain 
degree of fog is inevitable. 

Dec., 1945 



Owing to the fact that the active ingredients are not commercially 
available, they must be synthesized in the laboratory for experiments 
on the scale of preliminary tests such as these. However, the syn- 
thesis of the materials is relatively simple so that in the event the 
formula finds reasonable use in the industry the cost of the formula 
would be comparable to presently used commercial developers. 






10 O.O 


FIG. 6. 

A toxicity test was carried out by the author wherein a pad of 
cotton soaked in the developer solution was taped to the underside 
of the arm for a period of 14 hr. No effect of any kind was observed; 
not even a discoloration of the skin. There is, of course, the pos- 
sibility that certain individuals may be sensitive to the chemical con- 
tent of the Reidol developer. However, in view of the fact that the 
pH of the developer is near neutral and that the salt concentration is 
such that near isotonic solutions are obtained, it is reasonable to ex- 
pect that little difficulty will be had on this score. 

All H and D tests were made with Eastman Kodak Plus-X film. 


I ALBURGER, J. R.: "Mathematical Expression of Developer Behavior," /. 
Soc. Mot. Pict. Eng., XXXV, 3 (Sept., 1940), p. 282. 



Summary. Vari-focal view-finders now in use change the field which corresponds 
to the field covered by the camera lens either mechanically or optically. Mechanical 
systems change the area of the image seen through the view-finder and therefore produce 
small images when used with telephoto lenses. The optical view-finders now in use 
are either negative systems, which do not allow the image produced by them to be 
framed properly, or they also change the area of the image. 

The new view-finder consists of a vario-focal system with movable lenses producing 
real images corresponding to the different fields of the motion picture lenses at infinity, 
which are viewed through a stationary erector system . They therefore can be framed and 
the frame being inside the erector system always appears the same size regardless of 
the focal length of the view-finder. 

A new positive Vari-Focal view-finder for motion picture cameras 
eliminates the disadvantages of the 2 classes of finders now in use. 
The new view-finder uses only positive elements, thereby producing 
an upright real image. And it is possible to secure this real image in a 
frame of constant size. The image becomes variable over a wide 
range along a smooth and continuous curve instead of by intermittent 
steps as is the case in some other types of finders. 

The 2 classes of finders presently used secure images by 2 different 
methods. One limits the fields of the different camera lenses by 
mechanically changing the size of the frame which surrounds the 
image. The second class changes the field by optical means. 

In the first class, we see that the image frame gets smaller as the 
focal length of the camera objective increases. The disadvantages 
of this are apparent. Those finders of the second class which change 
the field optically do one of two things : they change the frame size, as 
in the first class, or they produce only virtual images which do not 
frame distinctly. Negative-type finders, as described, also have 

* Presented May 16, 1945, at the Technical Conference in Hollywood. 
**M.E., Sc.D., Research and Development Laboratory, 381 Fourth Ave., 
New York. 



FIG. 1, 


FIG. 2. 



Vol 45, No. 6 

large parallactic error. This is particularly true when the eye is dis- 
placed. One other type of view-finder having a turrethead might be 
mentioned here. Its evident disadvantage lies in the limitations 
placed upon it by the number of optical elements in its turrethead. 

The vari-focal system of the new positive view-finder is based, of 
course, on the principle of the astronomical telescope. The finder 
adds to this principle certain other features which give it decided ad- 
vantages. In the astronomical telescope, magnification is deter- 
mined by the ratio of the focal length of the front lens combination 
to the focal length of the rear lens combination. 

FIG. 3. 

Thus, a long focus front lens combined with a short focus rear lens 
gives us magnification. Fig. l(a) illustrates this clearly. Con- 
versely, a short focus front lens coupled with a long focus rear lens 
brings about reduction, as seen in Fig. l(b). 

The new positive view-finder uses a front lens and a rear lens of ap- 
proximately equal focal length. The front and rear lenses do not 
move and are known as ' 'stationary lenses." The variation is ob- 
tained by 2 lenses of shorter focal length mounted in a barrel within 
the view-finder housing between the 2 stationary lenses. These 2 
shorter focus or "variator lenses" can be moved forward and back- 

Dec., 1945 



ward from the front stationary lens to the rear lens. The movable 
barrel in which they are mounted is called the "variator." 

This is how the variator operates: The stationary, and variator 
lenses are computed in such a way that when the variator is in the ex- 
treme front position, the combined power of the stationary front lens 
and the first variator lens is such that the inverted real image pro- 
duced by them falls on the second variator lens. Fig. 2 (a) shows 
this. Thus the second variator lens acts as a field lens in this position 
and does not participate in the forming of the image. So far, 
the vari-focal system acts as a telescope that has a short focus objec- 
tive and a longer focus rear lens, producing a reduced image covering 
a wide-field angle, as with a wide-angle lens. 

FIG. 4. 

Magnification is produced as the variator barrel is moved toward 
the rear stationary lens. What happens here, of course, is that the 
combined focal length of the front stationary lens and the front vari- 
ator lens increases because the spacing between these 2 lenses be- 
comes wider. Simultaneously the rear variator lens begins to com- 
bine its power with that of the rear stationary lens. Fig. 2(b) shows 
what takes place here. By moving the variator barrel, the focal 
length of the combined front stationary lens and front variator lens 
becomes longer, while the focal length of the combined rear variator 
lens and rear stationary lens shortens as the spacing between them 
decreases. Thus, the change from a reduced image to an enlarged 
Dne occurs when the 'combined focal length of the front optical ele- 



Vol 45, No. 6 

ments becomes greater than the combined focal length of the rear 

The largest magnification is produced by the vari-focal system 
when, in the extreme rear position, the image formed by the front 
stationary lens falls into the front variator lens. In this position the 
vari-focal system covers only a small field angle and therefore corre- 
sponds to a telephoto lens . (See Fig. 2 (c) . ) 

Two serious disadvantages to the vari-focal system as a complete 
view-finder when used alone have been overcome in the new positive 

FIG. 5. 

view-finder. It has been noted, of course, that the objective lens 
combination produces an inverted image, an objectionable feature in a 
view-finder. A more serious defect of the "variator" system lies in 
the fact that the image moves in opposite direction to the movement 
of the 'Variator" barrel from the inside of the rear "variator lens" to 
the inside of the front "variator lens." So in spite of it being a real 
image, it cannot be framed. 

Combined with the vari-focal system to overcome these disadvan- 
tages is an erector system. Fig. 3 is a diagrammatic representation of 
the view-finder. The dotted line separates the vari-focal mechanism 
from the erector system. This erector system acts practically as a 


second telescope which collimates the image through the combination 
of the variator and stationary lenses. 

This new image is not only upright as the name erector system im- 
plies, but it remains practically stationary within the erector system, 
because the optical elements of this second telescope do not change. 
Therefore, a frame can be placed at the image point which corre- 
sponds to the size of the film frame and which shows accurately the 
image which will be produced by the corresponding camera lens. 

A cross section through the actual view-finder, illustrated in Fig. 
4, shows how the optical elements are arranged. How the new posi- 
tive view-finder appears mounted on the door of the Bell and Ho well 
Eyemo camera is shown in Fig. 5. 

472 ANNOUNCEMENTS Vol 45, No. 6 


The Society desires to compile a list of members who gave their lives while 
serving with the Armed Forces of their country. Such a list will include members 
abroad who served with Allied military forces as well as -those in the services of 
the United States. 

The general office of the Society is not always advised of deceased members, 
and it will be appreciated if readers of the JOURNAL will forward the name of any 
member known to them to have been a war casualty. Please include with names 
submitted the approximate date, place, and any other information available. 

Your cooperation will assist the general office in obtaining a complete and 
accurate list for the records of the Society. 



Designer and engineer experienced in optics, lighting, and microphotog- 
raphy, capable of designing microfilm reading equipment and products 
related to microfilm industry. Reply to Microstat Corporation, 18 
West 48th St., New York 19, N.Y. 

Design engineer, experienced in mechanics and optics of motion picture 
cameras, projectors, and film scanning. Give details. Reply to Mr. 
John H. Martin, Columbia Broadcasting System, Inc., 485 Madison 
Ave., New York 22, N.Y. 


Sound recording engineer, 16- or 35-mm equipment, studio or location 
work, single or double system. Free to travel. For details write J. J. K., 
354 Ninth Ave., New York 1, N.Y. 

Honorably discharged veteran with 15 years' experience in all phases of 
motion picture production, including film editing, directing, producing. For 
details write F. A., 30-71 34th St., Long Island City 3, N.Y. Telephone 
AStoria 8-0714. 

Projectionist-newsreel editor with 15 years' experience just released 
from service. Willing to locate anywhere. Write P. O. Box 152, Hamp- 
den Station, Baltimore 11, Maryland. 

Chief Engineer of motion picture camera manufacturer now available. 
Special training in optics, electricity, electronics, mechanics. Experienced 
in all phases of manufacture of cameras, projectors, and accessories. 
Prefer West Coast, but not essential. Write Robert Winkler, 119 West 
78th St., New York, N. Y, 









BACK, F. G. 
BIRD, L. F. 


(and SIMMONS, N. L., 
and HYNDMAN, D. E.) 


(and FRAYNE, J. G.) 




(and CRANE, G. R.) 

(and STROHM, J. T.) 


(and LANSING, J. B.) 


(and SIMMONS, N. L., 
and CORBIN, R. M.) 


(and ZAVESKY, R. J., 
and LOZIER, W. W.) 


Title No. Page 

A New Photographic Developer for 

Picture Negatives 6 (Dec.) 459 

Film The Backbone of Television 

Programming 6 (Dec.) 401 

A Positive Vari-Focal View-Finder 

for Motion Picture Cameras 6 (Dec.) 466 

An Automatic High-Pressure Mer- 
cury Arc Lamp Control Circuit 1 (July) 38 

Practical Utilization of Monopack 

Film 5 (Nov.) 327 

Two New Eastman Fine-Grain 

Sound Recording Films 4 (Oct.) 265 

Automatic Recording of Photo- 
graphic Densities 5 (Nov.) 370 

The Presentation of Technical De- 
velopments Before Professional 
Societies 3 (Sept.) 184 

Machine Processing of 16 -Mm An- 

sco Color Film 5 (Nov.) 313 

Automatic Recording of Photo- 
graphic Densities 5 (Nov.) 370 

The U. S. Naval Photographic Serv- 
ices Depot 4 (Oct.) 294 

Development of Two Automatic 
Follow-Focus Devices for Use in 
Cinematography 4 (Oct.) 302 

An Improved Loudspeaker System 

for Theaters 5 (Nov.) 339 

Two New Eastman Fine-Grain 
Sound Recording Films 4 (Oct.) 265 

Method for Measurement of Bright- 
ness of Carbon Arcs 1 (July) 10 

A New Carbon for Increased Light 

in Studio and Theater Projection 6 (Dec.) 449 





(and HILLIARD, J. K.) 


(and TWINING, S. J.) 
Lo, T. Y. 


(and JONES, M. T., 
and ZAVESKY, R. J.) 

(and ZAVESKY, R. J. 
and NULL, M. R.) 
McNAiR, J. W. 




(and RETTINGER, M.) 

NULL, M. R. 

(and ZAVESKY, R. J., 
and Lozier, W. W.) 

O'DEA, D. 


Title No. Page 

The Calculation of Accelerations in 
Cam-Operated Pull- Down Mecha- 
nisms 2 (Aug.) 143 

Efficiency of Picture Projection Sys- 
tems 3 (Sept.) 191 

An Improved Loudspeaker System 
for Theaters 5 (Nov.) 339 

Continuous Flash Lighting An 
Improved High-Intensity Light 
Source for High-Speed Motion 
Picture Photography 5 (Nov.) 358 

Variable-Area Release from Vari- 
able-Density Original Sound 
Tracks 5 (Nov.) 380 

Technical Problems of Interpreta- 
tion in Producing Foreign-Ver- 
sion Films 3 (Sept.) 203 

Method for Measurement of Bright- 
ness of Carbon Arcs 1 (July) 10 

A New Carbon for Increased Light 

in Studio and Theater Projection 6 (Dec.) 449 

Study of Radiant Energy at Motion 
Picture Film Aperture 2 (Aug.) 102 

Progress Report of the Work of the 
ASA War Committee on Photog- 
raphy and Cinematography, Z52 1 (July) 33 

Some Practical Aspects of the Inter- 
modulation Test 3 (Sept.) 161 

Preliminary Report of Academy Re- 
search Council Committee on Re- 
recording Methods for 16-Mm 
Release of 35-Mm Features 2 (Aug.) 135 

An Optical Cueing Device for Disk 

Playback 4 (Oct.) 297 

Du Pont Fine-Grain Sound Films- 
Types 232 and 236 4 (Oct.) 285 

Anecdotal History of Sound Re- 
cording Technique 1 (July) 48 

Study of Radiant Energy at Mo- 
tion Picture Film Aperture 2 (Aug.) 102 

Comparison of Variable- Area Sound 

Recording Films 1 (July) 1 

Some Notes on the Duplication of 

16-Mm Integral Tripack Color 

Films 2 (Aug.) 113 




(and MUELLER, W. A.) 

RYAN, B. F. 

(and SMITH, E. H.) 

(and SPARKS, M. R.) 


(and CORBIN, R. M., 
and HYNDMAN, D. E.) 

(and RYAN, B. F.) 

(and SHANER, V. C.) 


(and HECKLER, W. G.) 




(and LIVADARY, J. P.) 



(and JONES, M. T., 
and LOZIER, W. W.) 

(and NULL, M. R., 
and LOZIER, W. W.) 

Chambers for 

No. Page 



Anecdotal History of Sound Re- 

cording Technique 
Problems of Theater Television Pro- 

jection Equipment 
A Small Microphone Boom 
The Application of the Polarograph 

to the Analysis of Photographic 

Fixing Baths 

Two New Eastman Fine-Grain 
Sound Recording Films 

A Small Microphone Boom 

8000 Pictures Per Second 

The Application of the Polarograph 

to the Analysis of Photographic 

Fixing Baths 

Development of Two Automatic 

Follow-Focus Devices for Use in 

The Projection Life of Film 
Some Relationships Between the 

Physical Properties and the Be- 

havior of Motion Picture Film 
A Multisection Rerecording Equal- 

A 16-Mm Edge-Numbering Ma- 

A New Medium for the Production 

of Vandykes 
Variable- Area Release from Vari- 

able-Density Original Sound 


Power Rectifiers for Studio Lighting 
The Motion Picture and Interna- 

tional Enlightenment 
Method for Measurement of Bright- 

ness of Carbon Arcs 

5 (Nov.) 
1 (July) 

3 (Sept.) 

6 (Dec.) 

6 (Dec.) 
3 (Sept.) 

2 (Aug.) 
1 (July) 




1 (July) 20 

4 (Oct.) 265 


1 (July) 20 

4 (Oct.) 302 
2 (Aug.) 78 

3 (Sept.) 209 

5 (Nov.) 333 
2 (Aug.) 109 

1 (July) 54 

5 (Nov.) 380 

6 (Dec.) 414 


A New Carbon for Increased Light 

in Studio and Theater Projection 6 (Dec.) 449 
Study of Radiant Energy at Motion 

Picture Film Aperture 2 (Aug.) 102 



Academy of Motion Picture Arts and Sciences 
(See also Research Council} 

The Motion Picture and International Enlightenment, W. F. Wanger, No. 2 
(Aug.), p. 76. 


Reverberation Chambers for Rerecording, M. Rettinger, No. 5 (Nov.), p. 350. 

American Standards Association 

Progress Report of the Work of the ASA War Committee on Photography and 

Cinematography, Z52, J. W. McNair, No. 1 (July), p. 33. 


A 16-Mm Edge-Numbering Machine, L. Thompson, No. 2 (Aug.), p. 109. 

Problems of Theater Television Projection Equipment, A. H. Rosenthal, No. 3 
(Sept.), p. 218 

An Optical Cueing Device for Disk Playback, G. C. Misener, No. 4 (Oct.), p. 

Development of Two Automatic Follow-Focus Devices for Use in Cinema- 
tography, J. T. Strohm and W. G. Heckler, No. 4 (Oct.), p. 302. 

A Multisection Rerecording Equalizer, W. L. Thayer, No. 5 (Nov.), p. 333. 

Continuous Flash Lighting An Improved High-Intensity Light Source for 
High-Speed Motion Picture Photography, H. M. Lester, No. 5 (Nov.), p. 358. 

Automatic Recording of Photographic Densities, J. G. Frayne and G. R. Crane 
No. 5 (Nov.), p. 370 

Power Rectifiers for Studio Lighting, L. A. Umansky, No. 6 (Dec.), p. 414. 

A Small Microphone Boom, B. F. Ryan and E. H. Smith, No. 6 (Dec.), p. 441. 

A Positive Vari-Focal View-Finder for Motion Picture Cameras, F. G. Back, 
No. 6 (Dec.), p. 466. 


Method for Measurement of Brightness of Carbon Arcs, M. T. Jones, R. J. 

Zavesky, and W. W. Lozier, No. 1 (July) p. 10. 
An Automatic High-Pressure Mercury Arc Lamp Control Circuit, L. F. Bird, 

No. 1 (July), p. 38. 
Study of Radiant Energy at Motion Picture Film Aperture, R. J. Zavesky, 

M. R. Null, and W. W. Lozier, No. 2 (Aug.), p. 102. 
Studio Lighting Committee, Report of: (Operation and maintenance of studio 

lighting equipment, both carbon arc and incandescent) No. 4 (Oct.), p. 249. 
A New Carbon for Increased Light in Studio and Theater Projection, M. T. 

Jones, R. J. Zavesky, and W. W. Lozier, No. 6 (Dec.), p. 449. 


478 INDEX Vol 45, No. 6 

Atlantic Coast Section (See SMPE Activities and Announcements') 


The Calculation of Accelerations in Cam-Operated Pull-Down Mechanisms, 
E. W. Kellogg, No. 2 (Aug.), P- 143. 

Development of Two Automatic Follow-Focus Devices for Use in Cinematog- 
raphy, J. T. Strohm and W. G. Heckler, No. 4 (Oct.), p. 302. 

A Positive Van-Focal View-Finder for Motion Picture Cameras, F. G. Back, 
No. 6 (Dec.), p. 466. 


Development of Two Automatic Follow-Focus Devices for Use in Cinema- 
tography, J. T. Strohm and W. G. Heckler, No. 4 (Oct.), p. 302. 

Practical Utilization of Monopack Film, C. G. Clarke, No. 5 (Nov.), p. 327. 

A Positive Vari-Focal View-Finder for Motion Picture Cameras, F. G. Back, 
No. 6 (Dec.), p. 466. 

Cinematography, High-Speed 

8000 Pictures Per Second, H. J. Smith, No. 3 (Sept.), p. 171 
Continuous Flash Lighting An Improved High-Intensity Light Source for 
High-Speed Motion Picture Photography, H. M. Lester, No. 5 (Nov.), p. 358. 


Some Notes on the Duplication of 16-Mm Integral Tripack Color Film, W. H. 
Offenhauser, Jr., No. 2 (Aug.), p. 113. 

Machine Processing of 16-Mm Ansco Color Film, J. L. Forrest, No. 5 (Nov.), p. 

Practical Utilization of Monopack Film, C. G. Clarke, No. 5 (Nov.), p. 327. 

Color, Report of Committee on: (Discussion of new high sensitivity caesium- 
antimony phototubes) No. 6 (Dec.), p. 397. 

Committee Activities and Reports 

Color Committee: (Discussion of new high sensitivity caesium-antimony 
phototubes) No. 6 (Dec.), p. 397. 

Papers Committee: The Presentation of Technical Developments Before Pro- 
fessional Societies, W. L. Everitt, No. 3 (Sept.), p. 184. 

Projector Sprocket Design, Subcommittee on: (Results of tests on large-size pro- 
jector sprockets) No. 2 (Aug.), p. 73. 

Screen Brightness, Subcommittee on: (Projects under study) No. 4 (Oct.), 
p. 262. 

Standards Committee: (Projects under study) No. 4 (Oct.), p. 261. 

Theater Engineering, Construction, and Operation, Subcommittee on: (Formation 
of subcommittee to study various subjects relating to theaters; outline of 
project on theater carpets; study of theater building codes) No. 4 (Oct.), 
p. 262. 

Cueing Aids 

An Optical Cueing Device for Disk Playback, G. C. Misener, No. 4 (Oct.), p. 

Dec., 1945 INDEX 479 

Current Literature 

No. 1 (July), p. 65; No. 3 (Sept.), p. 241; No. 4 (Oct.), p. 310. 


Automatic Recording of Photographic Densities, J. G. Frayne and G. R. Crane, 
No. 5 (Nov.), p. 370. 

Developing (See Processing} 


Technical Problems of Interpretation, in Producing Foreign- Version Films, 
T. Y. Lo, t No. 3 (Sept.), p. 203. 

Edge Damage (See Film Wear] 
Edge-Numbering (See Editing) 

A 16-Mm Edge- Numbering Machine, L. Thompson, No. 2 (Aug.), p. 109. 

Electronic Tubes 

Color, Report of Committee on: (Discussion of new high sensitivity caesium- 
antimony phototubes) No. 6 (Dec.), p. 397. 

Federal Communications Commission (See Television} 

Fellow Membership Awards (See SMPE Activities and Announcements') 

Film, Distortion 

Some Relationships Between the Physical Properties and the Behavior of 
Motion Picture Film, R. H. Talbot, No. 3 (Sept.), p. 209. 

Film, Fine-Grain 

Comparison of Variable-Area Sound Recording Films, D. O'Dea, No. 1 (July) 

P. 1. 
Two New Eastman Fine-Grain Sound Recording Films, R. M. Corbin, N. L. 

Simmons, and D. E. Hyndman, No. 4 (Oct.), p. 265. 
Du Pont Fine-Grain Sound Films Types 232 and 236, H. W. Moyse, No. 4 

(Oct.), p. 285. 

Film, General 

A 16-Mm Edge-Numbering Machine, L. Thompson, No. 2 (Aug.), p. 109. 
Some Relationship Between the Physical Properties and the Behavior of Motion 

Picture Film, R. H. Talbot, No. 3 (Sept.), p. 209. 
Film The Backbone of Television Programming, R. B. Austrian, No. 6 

(Dec.), p. 401. 

Film Wear 
The Projection Life of Film, R. H. Talbot, No. 2 (Aug.), p. 78. 

Fixing Baths 

The Application of the Polarograph to the Analysis of Photographic Fixing 
Baths, V. C. Shaner and M. R. Sparks, No. 1 (July), p. 20. 

480 INDEX Vol 45, No. 6 


A New Medium for the Production of Vandykes, L. S. Trimble, No. 1 (July), 

p. 54. 
The Motion Picture and International Enlightenment, W. F. Wanger, No. 2 

(Aug.), p. 76. 
The Presentation of Technical Developments Before Professional Societies, 

W. L. Everitt, No. 3 (Sept.), p. 184. 
Technical Problems of Interpretation in Producing Foreign-Version Films, 

T. Y. Lo, No. 3 (Sept.), p. 203. 

High-Speed Photography (See Cinematography, High Speed} 


Method for Measurement of Brightness of Carbon Arcs, M. T. Jones, R. J. 

Zavesky, and W. W. Lozier, No. 1 (July), p. 10. 
An Automatic High-Pressure Mercury Arc Lamp Control Circuit, L. F. Bird, 

No. 1 (July), p. 38. 
Study of Radiant Energy at Motion Picture Film Aperture, R. J. Zavesky, 

M. R. Null, and W. W. Lozier, No. 2 (Aug.), p. 102. 

Efficiency of Picture Projection Systems, E. W. Kellogg, No. 3 (Sept.), p. 191. 
Studio Lighting Committee, Report of: (Operation and maintenance of studio 

lighting equipment, both carbon arc and incandescent) No. 4 (Oct.), p. 249. 
Screen Brightness, Report of Subcommittee on: (Projects understudy) No. 4 

(Oct.), p. 262. 
Continuous Flash Lighting An Improved High-Intensity Light Source for 

High-Speed Motion Picture Photography, H. M. Lester, No. 5 (Nov.), p. 358. 
Power Rectifiers for Studio Lighting, L. A. Umansky, No. 6 (Dec.), P- 414. 
A New Carbon for Increased Light in Studio and Theater Projection, M. T. 

Jones, R. J. Zavesky, and W. W. Lozier, No. 6 (Dec.), p. 449. 

Incandescent Lamps 

Studio Lighting Committee, Report of: (Operation and maintenance of studio 
lighting equipment, both carbon arc and incandescent) No. 4 (Oct.), p. 249. 


The Application of the Polarograph to the Analysis of Photographic Fixing 
Baths, V. C. Shaner and M. R. Sparks, No. 1 (July), p. 20. 

Journal Award (See SMPE Activities and Announcements') 

Laboratory Practice 

Variable-Area Release from Variable-Density Original Sound Tracks, J. P. 

Livadary and S. J. Twining, No. 5 (Nov.), p. 380. 
A 16-Mm Edge-Numbering Machine, L. Thompson, No. 2 (Aug.), p. 109. 


An Improved Loudspeaker System for Theaters, J. B. Lansing and J. K. 
Hilliard, No. 5 (Nov.), p. 339. 

Mercury Arcs 

An Automatic High-Pressure Mercury Arc Lamp Control Circuit, L. F. Bird, 
No. 1 (July), p. 38. 

Dec., 1945 INDEX 481 


Anecdotal History of Sound Recording Technique, W. A. Mueller and M. 

Rettinger, No. 1 (July), p. 48. 
A Small Microphone Boom, B. F. Ryan and E. H. Smith, No. 6 (Dec.), p. 441. 


C. E. Shultz, No. 2 (Aug.), p. 159. 


Efficiency of Picture Projection Systems, E. W. Kellogg, No. 3 (Sept.), p. 191. 

Pacific Coast Section (See SMPE Activities and Announcements) 
Phototubes (See Electronic Tubes) 

Process Photography 

Technical News, No. 2 (Aug.), p. 156. 

A New Carbon for Increased Light in Studio and Theater Projection, M. T. 
Jones, R. J. Zavesky, and W. W. Lozier, No. 6 (Dec.), p. 449. 


The Application of the Polarograph to the Analysis of Photographic Fixing 

Baths, V. C. Shaner and M. R. Sparks, No. 1 (July), p. 20. 
Some Notes on the Duplication of 16-Mm Integral Tripack Color Films, W. H. 

Offenhauser, Jr., No. 2 (Aug.), p. 113. 
Some Practical Aspects of the Intermodulation Test, E. Meschter, No. 3 (Sept.), 

p. 161. 
Some Relationships Between the Physical Properties and the Behavior of 

Motion Picture Film, R. H. Talbot, No. 3 (Sept.), p.' 209. 
Machine Processing of 16-Mm Ansco Color Film, J. L. Forrest, No. 5 (Nov.), 

p. 313. 
Variable-Area Release from Variable-Density Original Sound Tracks, J. P. 

Livadary and S. J. Twining, No. 5 (Nov.), p. 380. 
A New Photographic Developer for Picture Negatives, J. R. Alburger, No. 6 

(Dec.), p. 459. 


Technical Problems of Interpretation in Producing Foreign-Version Films, 
T. Y. Lo, No. 3 (Sept.), p. 203. 

The U. S. Naval Photographic Services Depot, F. M. Hearon, No. 4 (Oct.), p. 

An Optical Cueing Device for Disk Playback, G. C. Misener, No. 4 (Oct.), p. 

Development of Two Automatic Fo'llow-Focus Devices for Use in Cinema- 
tography, J. T. Strohm and W. G. Heckler, No. 4 (Oct.), p. 302. 

Practical Utilization of Monopack Film, C. G. Clarke, No. 5 (Nov.), p. 327. 

Power Rectifiers for Studio Lighting, L. A. Umansky, No. 6 (Dec.), p. 414. 

A Small Microphone Boom, B. F. Ryan and E. H. Smith, No. 6 (Dec.), p. 441. 


The Projection Life of Film, R. H. Talbot, No. 2 (Aug.), p. 78. 

482 INDEX Vol 45, No. 6 

Study of Radiant Energy at Motion Picture Film Aperture, R. J. Zavesky, 

M. R. Null, and W. W. Lozier, No. 2 (Aug.), p. 102. 
Efficiency of Picture Projection Systems, E. W. Kellogg, No. 3 (Sept.), p. 191. 


Projector Sprocket Design, Report of Subcommittee on: (Results of tests on 

large-size projector sprockets) No. 2 (Aug.), p. 73. 

The Calculation of Accelerations in Cam-Operated Pull-Down Mechanisms, 
E. W. Kellogg, No. 2 (Aug.), p. 143. 

Pull-Down Mechanisms 

The Calculation of Accelerations in Cam-Operated Pull-Down Mechanisms, 
E. W. Kellogg, No. 2 (Aug.), p. 143. 


Power Rectifiers for Studio Lighting, L. A. Umansky, No. 6 (Dec.), p. 414. 

Rerecording (See Sound Recording} 

Research Council, Academy of Motion Picture Arts and Sciences 
(See also Academy of Motion Picture Arts and Sciences} 
Preliminary Report of Academy Research Council Committee on Rerecording 

Methods for 16-Mm Release of 35-Mm Features, W. C. Miller, No. 2 (Aug.), 

p. 135. 

Screen Brightness (See Illumination) 


Automatic Recording of Photographic Densities, J. G. Frayne and G. R. Crane, 
No. 5 (Nov.), p. 370. 

Sixteen-Mm Motion Pictures 

A 16-Mm Edge-Numbering Machine, L. Thompson, No. 2 (Aug.), p. 109. 
Some Notes on the Duplication of 16-Mm Integral Tripack Color Films, W. H. 

Offenhauser, Jr., No. 2 (Aug.), p. 113. 
Preliminary Report of Academy Research Council Committee on Rerecording 

Methods for 16-Mm Release of 35-Mm Features, W. C. Miller, No. 2 (Aug.), 

p. 135. 
Some Practical Aspects of the Intermodulation Test, E. Meschter, No. 3 

(Sept.), p. 161. 
Machine Processing of 16-Mm Ansco Color Film J. L. Forrest, No. 5 (Nov.), 

p. 313. 

SMPE Activities and Announcements 
(See also Committee Activities and Reports') 

Amendments of By-Laws, No. 3 (Sept.), p. 246; No. 5 (Nov.), p. 389. 
Atlantic Coast Section : 

Meetings, May 23 and June 13 No. 1 (July), p. 67. 

Officers.and Managers for 1946-47, No. 5 (Nov.), p. 390. 
Fellow Membership Awards: Announcement of 1945 elections, No. 5 (Nov.), 

p. 389. 

Dec., 1945 INDEX 483 

Fifty-Seventh Semi-Annual Technical Conference: Final Program, No. 1 

(July), p. 67. 
The Motion Picture and International Enlightenment, W. F. Wanger, No. 2 

(Aug.), P. 76. 
Fifty-Eighth Semi-Annual Technical Conference: Committees and Tentative 

Program, No. 3 (Sept.), P- 243. 
Final Program, No. 5 (Nov.), p. 391. 

Journal Award: Announcement of 1945 recipients, No. 5 (Nov.), p. 389. 
Members Lost Serving Their Country, To Compile List of, No. 6 (Dec.), p. 472. 
Officers, Governors, and Section Managers for 1946-47, No. 5 (Nov.), p. 390. 
Pacific Coast Section : 

Meeting, June 26 No. 2 (Aug.), p. 158; Meeting, September 25 No. 5 
(Nov.), P. 390. 

Officers and Managers for 1946-47, No. 5 (Nov.), p. 390. 

Sound Effects 

Reverberation Chambers for Rerecording, M. Rettinger, No. 5 (Nov.), P. 350. 

Sound Recording 

Comparison of Variable- Area Sound Recording Films, D. O'Dea, No. 1 (July), 
p. 1. 

Anecdotal History of Sound Recording Technique, W. A. Mueller and M. Ret- 
tinger, No. 1 (July), p. 48. 

Preliminary Report of Academy Research Council Committee on Rerecording 
Methods for 16-Mm Release of 35-Mm Features, W. C. Miller, No. 2 (Aug.), 
p. 135. 

Some Practical Aspects of the Intel-modulation Test, E. Meschter, No. 3 (Sept.), 
p. 161. 

Two New Eastman Fine-Grain Sound Recording Films, R. M. Corbin, N. L. 
Simmons, and D. E. Hyndman, No. 4 (Oct.), p. 265. 

Du Pont Fine-Grain Sound Films Types 232 and 236, H. W. Moyse, No. 4 
(Oct.), p. 285. 

A Multisection Rerecording Equalizer, W. L. Thayer, No. 5 (Nov.), p. 333. 

Reverberation Chambers for Rerecording, M. Rettinger, No. 5 (Nov.), p. 350. 

Variable-Area Release from Variable-Density Original Sound Tracks, J. P. 
Livadary and S. J. Twining, No. 5 (Nov.), p. 380. 

Color, Report of Committee on: (Discussion of new high sensitivity caesium- 
antimony phototubes) No. 6 (Dec.), p. 397. 

Sound Reproduction 

Some Practical Aspects of the Intermodulation Test, E. Meschter, No. 3 

(Sept.), p. 161. 
An Improved Loudspeaker System for Theaters, J. B. Lansing and J. K. 

Hilliard, No. 5 (Nov.), p. 339. 


Progress Report of the Work of the ASA War Committee on Photography and 

Cinematography, Z52, J. W. McNair, No. 1 (July), p. 33. 
Standards Committee, Report of: (Projects under study) No. 4 (Oct.), p. 261. 

484 INDEX 

Technical News 

No. 2 (Aug.), P. 156. 


Frequency Allocations for Theater Television: Excerpts from Report by 

Federal Communications Commission on Proposed Allocations from 25,000 

Kilocycles to 30,000,000 Kilocycles, No. 1 (July), p. 16. 
Technical News, No. 2 (Aug.), p. 156. 
Problems of Theater Television Projection Equipment, A. H. Rosen thai, No. 3 

(Sept.), p. 218. 
Film The Backbone of Television Programming, R. B. Austrian, No. 6 (Dec.), 

p. 401. 

Test Films 

Preliminary Report of Academy Research Council Committee on Rerecording 
Methods for 16-Mm Release of 35-Mm Features, W. C. Miller, No. 2 (Aug.), 
p. 135. 

Theaters, General 

Theater Engineering, Construction, and Operation, Report of Subcommittee 
on: (Formation of subcommittee to study various subjects relating to 
'theaters; outline of project on theater carpets; study of theater building 
codes) No. 4 (Oct.), p. 262. 

An Improved Loudspeaker System for Theaters, J. B. Lansing and J. K. 
Billiard, No. 5 (Nov.), p. 339. 

Training Films 

The U. S. Naval Photographic Services Depot, F. M. Hearon, No. 4 (Oct.), p. 

U. S. Navy (See Training Films) 

War Committee on Photography and Cinematography, Z52 

Progress Report of the Work of the ASA War Committee on Photography and 

Cinematography, Z52, J. W. McNair, No. 1 (July), p. 33. 
Preliminary Report of Academy Research Committee on Rerecording Methods 

for 16-Mm Release of 35-Mm Features, W. C. Miller, No. 2 (Aug.), p. 135.